business27eee.com

BIGDATAANDTHE2030AGENDAFORSUSTAINABLEDEVELOPMENTBYABBASMAAROOF,PhD1TableofContentsAcknowledgementsExecutiveSummary1.
Introduction.
82.
BigDataandtheDataRevolution.
112.
1.
Fromthe3Vstothe3CsofBigData.
112.
2.
UseofBigData.
142.
3.
BigDataandOpenData.
153.
TheBigDataEcosystem.
173.
1.
TypesofStakeholders.
173.
2.
RolesofStakeholdersintheEcosystem.
184.
BigDataandPolicyMaking214.
1.
BestPractices.
214.
2ThePolicyCycle.
225.
ChallengesandOpportunities.
265.
1.
Challenges.
265.
2.
Opportunities.
286.
BigDataandPolicyforthe2030AgendaforSustainableDevelopment.
306.
1.
AVisionforBigDataandthe2030Agenda.
306.
2.
PossibleAction.
31ListofTables,FiguresandBoxesTable1:RolesofStakeholdersinDataEcosystemFigure1:BigDataandPolicyFigure2:The3V'sFigure3:The3C's"DiagramFigure4:TheInterfaceofBigDataandOpenDataFigure5:Policy-MakingProcessBox1:Real-worldpolicyconstraints:theODIsurveyBox2:TwitterExample:UseofMobileTechnologyforPerceptionAssessmentBox3:SIMGovernmentBox4:AgilePredictivePolicyAnalysis(APPA)Annex1:BigDataTypesAnnex2:BigDataCaseStudiesAnnex3:Expertcomments,inputs,andrecommendationstoESCAPDraftReportonBigDataandthe2030AgendaforSustainableDevelopment2AcknowledgementsTheauthorwouldliketogratefullyacknowledgeadvicefromtheUN-ESCAPEnvironmentandDevelopmentPolicySection,especiallyHalaRazian.
ThisreportisaresultofcollaborationbetweenUNESCAP,EnvironmentandDevelopmentPolicySection,andtheUNGlobalPulseLabJakartaoffice.
TheauthorbenefittedfrominputsfromthePulseLabJakartateam.
IwouldalsothanktheUNGlobalPulseLabJakartaofficeforprovidingspaceandresourcesduringthedeskresearch.
TheauthorisalsogratefulforinputsandcommentsintothedraftreportbyKatinkaWeinberger,RusyanJillMamiitandMarkoJavorsek.
Thefindings,interpretations,andconclusionsexpressedinthisreportsolelyreflecttheauthor'sviews.
3GlossaryAlgorithm-Aformulaorstep-by-stepprocedureforsolvingaproblem.
Anonymization-Theprocessofremovingspecificidentifiers(oftenpersonalinformation)fromadataset.
Bigdata-Atermforalargedataset.
Bigdataanalytics-Atypeofquantitativeresearchthatexamineslargeamountsofdatatouncoverhiddenpatterns,unknowncorrelationsandotherusefulinformation.
Bigdatafordevelopment-AconceptthatreferstotheidentificationofsourcesofBigDatarelevanttopolicyandplanningofdevelopmentprograms.
Citizenreportingorcrowd-sourceddata-Informationactivelyproducedorsubmittedbycitizensthroughmobilephone-basedsurveys,hotlines,user-generatedmaps,etc;whilenotpassivelyproduced,thisisakeyinformationsourceforverificationandfeedbackDataexhaust-Passivelycollectedtransactionaldatafrompeople'suseofdigitalserviceslikemobilephones,purchases,websearches,etc.
,thesedigitalservicescreatenetworkedsensorsofhumanbehavior.
Datamining-Atermreferstotheactivityofgoingthroughbigdatasetstolookforrelevantorpertinentinformation.
Dataphilanthropy-Atermthatdescribesanewformofpartnershipinwhichprivatesectorcompaniessharedataforpublicbenefit.
Datacleaning/cleansing-Thedetectionandremoval,orcorrection,ofinaccuraterecordsinadataset.
Datamigration-Thetransitionofdatafromoneformatorsystemtoanother.
Datascience-Thegleaningofknowledgefromdataasadisciplinethatincludeselementsofprogramming,mathematics,modeling,engineeringandvisualization.
Datasilos-FixedorisolateddatarepositoriesthatdonotinteractdynamicallywithothersystemsExabyte-Alargeunitofcomputerdatastorage,twotothesixtiethpowerbytes.
Theprefixexameansonebillion,oronequintillion.
Indecimalterms,anexabyteisabilliongigabytes.
4Geospatialanalysis-Aformofdatavisualizationthatoverlaysdataonmapstofacilitatebetterunderstandingofthedata.
Realtimedata-Adatathatcovers/isrelevanttoarelativelyshortandrecentperiodoftime-suchastheaveragepriceofacommodityoverafewdaysratherthanafewweeks,andismadeavailablewithintimeframethatallowsactiontobetakenthatmayaffecttheconditionsreflectedinthedata.
Infographic-Agraphicvisualrepresentationsofinformation,dataorknowledgeintendedtopresentinformationquicklyandclearlyMashup-Theuseofdatafrommorethanonesourcetogeneratenewinsight.
Statusquo-Atermreferstotheexistingstateofaffairs,particularlywithregardstosocialorpoliticalissues.
Opendata-Atermreferstodatathatisfreefromcopyrightandcanbesharedinthepublicdomain.
Openwebdata-Webcontentsuchasnewsmediaandsocialmediainteractions(e.
g.
blogs,Twitter),newsarticlesobituaries,e-commerce,jobpostings;sensorofhumanintent,sentiments,perceptions.
Onlineinformation-Webcontentsuchasnewsmediaandsocialmediainteractions(e.
g.
blogs,Twitter),newsarticlesobituaries,e-commerce,jobpostings;thisapproachconsiderswebusageandcontentasasensorofhumanintent,sentiments,perceptions,andwant.
Petabyte-Ameasureofmemoryorstoragecapacityandis2tothe50thpowerbytesor,indecimal,approximatelyathousandterabytes.
Predictiveanalytics/modeling-Theanalysisofcontemporaryandhistorictrendsusingdataandmodelingtopredictfutureoccurrences.
Physicalsensors-Satelliteorinfraredimageryofchanginglandscapes,trafficpatterns,lightemissions,urbandevelopmentandtopographicchanges,etc.
;remotesensingofchangesinhumanactivity.
Quantitativedataanalysis-Theuseofcomplexmathematicalorstatisticalmodelingtoexplain,orpredict,financialandbusinessbehavior.
Sentimentanalysis(opinionmining)-Theuseoftextanalysisandnaturallanguageprocessingtoassesstheattitudesofaspeakerorauthor,oragroup.
5Structureddata-Dataarrangedinanorganizeddatamodel,likeaspreadsheetorrelationaldatabase.
Semantics-Atermreferstothestudyofmeaning.
Itfocusesontherelationbetweensignifiers,likewords,phrases,signs,andsymbols,andwhattheystandfor;theirdenotation.
Tweet-ApostviatheTwittersocialnetworkingsiterestrictedtoastringupto140charactersUnstructureddata-Datathatcannotbestoredinarelationaldatabaseandcanbemorechallengingtoanalyzefromdocumentsandtweetstophotosandvideos.
6ExecutiveSummaryThisstocktakingreportattemptstoprovideanoverviewofbigdata,itsuseinthepolicy-makingcontext,thestakeholdersandtheirrolesandprovidessomesuggestedactionablestepsasadiscussionstimulusforthe"BigDataandthe2030AgendaforSustainableDevelopment:AchievingtheDevelopmentGoalsintheAsiaandthePacificRegion"meetinginBangkokon14-15December2015.
Criticaldataforglobal,regionalandnationaldevelopmentpolicymakingarestilllacking.
Manygovernmentsstilldonothaveaccesstoadequatedataontheirentirepopulations.
Thisisparticularlytrueforthepoorestandmostmarginalized,theverypeoplethatleaderswillneedtofocusoniftheyaretoachievezeroextremepovertyandzeroemissions,andto'leavenoonebehind'inthenext15years.
Thisistrue,too,fortheinternationalcommunity,whowillnotbeabletosupportthemostvulnerableandmarginalizedpeoplewithoutanoverhaulofthecurrentwaysofgatheringdata.
Whilemostdataistechnically"public",accessingitisnotalwayseasy,andminingitforrelevantinsightscanrequiretechnicalexpertiseandtrainingthatorganizationsandgovernmentswithlimitedresourcescan'talwaysafford.
Makinggooduseofbigdatawillrequirecollaborationofvariousactorsincludingdatascientistsandpractitioners,leveragingtheirstrengthstounderstandthetechnicalpossibilitiesaswellasthecontextwithinwhichinsightscanbepracticallyimplemented.
RecentdiscussionsuggeststomoveawayfromseeingBigDatainisolation,buttoratherfocusonthe"ecosystem"ofBigData.
Accordingtothisconcept,BigDataisnotjustdata—nomatterhowbigordifferentitisconsideredtobe;bigdataisfirstandforemost'about'theanalytics,thetoolsandmethodsthatareusedtoyieldinsights,theframeworks,standards,stakeholdersinvolvedandthen,knowledge.
EffectiveapplicationofBigDatawouldalsorequirechangesinthedecision-makingprocess,whichcustomarilyreliesontraditionalstatistics.
GiventhehighfrequencyofBigData,amoreresponsivemechanismwillneedtobeputinplacethatallowsthegovernmenttoprocesstheinformationandactquicklyinresponse.
However,thisstocktakefindsthatbigdataisnot(yet)playingacrucialroleinpolicymaking.
Ifatall,itisusedattheagendasettingstageand/orevaluationstageofpolicymaking.
Oneofthereasonsmightbebecausetheecosystemisnotyetfunctioningandcrucialelements,suchasstandardsandframeworksarestillmissing.
Nationalgovernmentsandotherpolicymakersarejuststartingtosystematicallyengagewithbigdataforpolicymaking.
TheproposedstepsarebasedontherecommendationsoftheUNIndependentAdvisoryGroup,andaremeanttohelpbuildingandmaintainingtheBigDataecosystemforbetterdevelopmentpolicymaking:EstablishandmanageacoordinationmechanismwiththekeyUNstakeholdersandotherinternationalpartners;7DevelopaconsensusonprinciplesandstandardsamongtheUNESCAPmembercountries;Kick-offandinstitutionalizeaRegionalMulti-StakeholderMechanismtoshareinnovations;MobilizeregionalresourcesforcapacitydevelopmentforthelessadvancedUNESCAPmembercountries;Enhancein-housebigdataanalyticscapacity.
Dependingonthediscussionsduringtheworkshopandagreementsbetweenstakeholders,certainrecommendedactionscouldbeprioritizedandelaboratedfurther.
81.
IntroductionBigdataapplicationsmayoffertheabilitytocollectandanalyze'realtime'informationfromacrossESCAP's62memberStatesforpoliciesthatrelatetothe2030Agenda's17goalsandtheir169targets.
Thescopeofthisinformationisvast,andbigdataapplicationscanfacilitatepolicymakingintheregionthatwouldotherwiserequirededicatedintensiveandcontinuoushumanandfinancialresources.
Thisstocktakingreport,commissionedbyESCAP,attemptstoprovideanoverviewofbigdata,itsuseinthepolicy-makingcontext,thestakeholdersandtheirrolesinmakingthemostoutoftheopportunitiesthatbigdatapresents.
Forillustrativepurposes,thereportthenpresentsaselectionofbestpracticesusingbigdatainthepolicymakingprocess.
Thereportthenwill,builtonexistingworkinthisfield,providesomepracticalideasonhowtofurtherprogressthe2030Agendaandpolicymakingarounditusingbigdata.
TherecommendationsofthisreportalsoshallinformESCAP'sstrategicplanningforthedevelopmentoftargetedcapacitybuildingprogramactivities,andtheAsiaPacificSustainableDevelopmentRoadmap.
Thediscussionofbigdataisquitecomplex,rangingfrompracticalortechnicalchallengestolegalandregulatorylimitations.
Thebelowfigure(Figure1)illustratesthe3differentdimensionsofbigdataandpolicies.
Whilethisreporttouchesonthepolicyfordatainthegapsandconstraintssection,thefocusofthisreportismainlyontheinnercircle:dataforpolicy.
Thecasestudiescomplementthecenterpiece-evidenceinformedpolicy-making.
ThepurposeofthisreportistosupportESCAP'sworkofprovidingrigorousanalysisandpeerlearning;andtranslatingthesefindingsintopolicydialoguesandrecommendations.
Itfocusesonbigdatainthepolicycontextandinthecontextofthe2030Agenda.
Despiteimprovements,criticaldataforglobal,regionalandnationaldevelopmentpolicymakingarestilllacking.
Largedatagapsremaininseveraldevelopmentareas.
Poordataquality,lackoftimelydataandunavailabilityofdisaggregateddataonimportantdimensionsareamongthemajorchallenges.
Asmanyas350millionpeopleworldwidearenotcoveredbyhouseholdsurveys.
Therecouldbeasmanyasaquartermorepeoplelivingonlessthan$1.
25adaythancurrentestimatessuggest,becausetheyhavebeenmissedoutofofficialsurveys[1].
9Figure1:BigDataandPolicyAsaresult,manynationalandlocalgovernmentscontinuetorelyonoutdateddataordataofinsufficientqualitytomakeplanninganddecisions.
Goodquality,relevant,accessibleandtimelydataenablesgovernmentstoextendtargetedservicesintocommunities,andtoimplementpoliciesmoreefficiently.
Manygovernmentsstilldonothaveaccesstoadequatedataontheirentirepopulations,andparticularlytrueforthepoorestandmostmarginalized,theverypeoplethatleaderswillneedtofocusoniftheyareto'leavenoonebehind'inthenext15years[2].
Thisistrue,too,fortheinternationalcommunity,whowillnotbeabletosupportthemostvulnerableandmarginalizedpeoplewithoutanoverhaulofthecurrentwaysofgatheringdata.
Box1:Real-worldpolicyconstraints:theODIsurveyToconfirmsomeoftheanecdotalevidenceaboutthelackofgooddataindevelopingcountryministries,theOverseasDevelopmentInstitute(ODI)interviewedaseriesofpolicy-makersbasedinlineministriestounderstandhowtheyviewedcapacityconstraintsintheirrespectivecountries.
Findingshighlightedtheproblemswithstabilityandcontinuityofdatacollection,particularlyincountriesinconflictwhereoftendataandinstitutionalmemoryarelostduringthewar,impactingtime-seriesanalysis.
Afurtherchallengewasmorepoliticalinnature,especiallyaroundalimitedunderstandingofhowthepublicsector10andcivilservantscanworkwithdataandhowdataservesthem,whichmaycauseresistancetoutilizationofdataeffectively.
Politicalissuesaresometimesmisconstruedbydevelopmentactorsascapacityissues[3].
Dataarenotjustaboutmeasuringchanges;theyalsofacilitateandcatalyzethatchange.
Ofcourse,goodqualitynumberswillnotchangepeople'slivesinthemselves.
Buttotargetthepoorestsystematically,toliftandkeepthemoutofpoverty,eventhemostwillinggovernmentscannotefficientlydeliverservicesiftheydonotknowwhothosepeopleare,wheretheyliveandwhattheyneed.
Nordotheyknowwheretheirresourceswillhavethegreatestimpact.
Policy-makingtakesplaceinanincreasinglyrichdataenvironment,whichposesbothpromisesandchallengestopolicy-makers.
Dataoffersachanceforpolicy-makingandimplementationtobemorecitizen-focused,takingaccountofcitizens'needs,preferencesandactualexperienceofpublicservices,asrecordedonsocialmediaplatforms.
AscitizensexpresspolicyopinionsonsocialnetworkingsitessuchasTwitterandFacebook;rateorrankservicesoragenciesongovernmentapplications;orenterdiscussionsonarangeofsocialenterpriseandNGOsites,theygenerateawholerangeofdatathatgovernmentagenciesmightharvesttogooduse.
Policy-makersalsohaveaccesstoahugerangeofdataoncitizens'actualbehaviour,asrecordeddigitallywhenevercitizensinteractwithgovernmentadministrationorundertakesomeactofcivicengagement,suchassigningapetition.
Dataminedfromsocialmediaoradministrativeoperationsinthiswayalsoprovidearangeofnewdata,whichcanenablegovernmentagenciestomonitor-andimprove-theirownperformance,forexamplethroughlogusagedataoftheirownelectronicpresenceortransactionsrecordedoninternalinformationsystems,whichareincreasinglyinterlinked.
Governmentscanusedatafromsocialmediaforself-improvement,byunderstandingwhatpeoplearesayingaboutgovernment,andwhichpolicies,servicesorprovidersareattractingnegativeopinionsandcomplaints,enablingidentificationofafailingschool,hospitalorcontractor,forexample.
Theycansolicitsuchdataviatheirownsites,orthoseofsocialenterprises.
Andtheycanfindoutwhatpeopleareconcernedaboutorlookingfor,fromtheGoogleSearchAPIorGoogletrends,whichrecordthesearchpatternsofahugeproportionofInternetusers[4].
TherecentreportoftheUNSecretaryGeneral'sIndependentExpertAdvisoryGroup(IEAG)[5]"definesthedatarevolutionforsustainabledevelopmentastheintegrationofdatacomingfromnewtechnologieswithtraditionaldatainordertoproducerelevanthigh-qualityinformationwithmoredetailsandathigherfrequenciestofosterandmonitorsustainabledevelopment.
Thisrevolutionalsoentailstheincreaseinaccessibilitytodatathroughmuchmoreopennessandtransparency,andultimatelymoreempoweredpeopleforbetterpolicies,betterdecisionsandgreaterparticipationandaccountability,leadingtobetteroutcomesforthepeopleandtheplanet".
112.
BigDataandtheDataRevolutionBigDataisnotasingle'thing'-itisacollectionofdatasources,technologiesandmethodologiesthathaveemergedfrom,andto,exploittheexponentialgrowthindatacreationoverthepastdecade[6].
Bigdataisabuzzword;usedtodescribeamassivevolumeofbothstructuredandunstructureddatathatissolargeitisdifficulttoprocessusingtraditionaldatabaseandsoftwaretechniques.
Dataisagrowingelementofourlives.
Moreandmoredataisbeingproducedandbecomingknowninthepopularliteratureas"bigdata",itsusageisbecomingmorepervasive,anditspotentialforpolicymakingandinternationaldevelopmentisjustbeginningtobeexplored[7].
2.
1.
Fromthe3Vstothe3CsofBigDataBigdatacanbedefinedaslargevolumesofhighvelocity,complex,andvariabledatathatrequireadvancedtechniquesandtechnologiestoenablethecapture,storage,distribution,managementandanalysisoftheinformation.
Bigdatacanbecharacterizedby3Vs:theextremevolumeofdata,thewidevarietyoftypesofdataandthevelocityatwhichthedatacanbeprocessed[8,9,10].
Althoughbigdatadoesn'trefertoanyspecificquantity,thetermisoftenusedwhenspeakingaboutpetabytesandexabytesofdata,muchofwhichcannotbeintegratedeasily.
Itisworthtomentionthatmostrecently,somedatascientistsandresearchershaveintroducedafourthcharacteristic,veracity,or'dataassurance'.
Thatis,thebigdataanalyticsandoutcomesareerror-freeandcredible.
However,veracityisstillagoalandnot(yet)areality[8].
Annex1describesthemostcommontypesofbigdata.
12Figure2:The3V's[11]Datasetsgrowinsizeinpartbecausetheyareincreasinglybeinggatheredbyinexpensiveandnumerousinformation-sensing,mobiledevices,remotesensing,softwarelogs,cameras,microphones,radio-frequencyidentification(RFID)readers,andwirelesssensornetworks[12],[13],[14].
Theworld'stechnologicalper-capitacapacitytostoreinformationhasroughlydoubledevery40monthssincethe1980s[15];asof2012,everyday2.
5exabytes(2.
5*1018)ofdatawerecreated.
Asof2014,everyday2.
3zettabytes(2.
3*1021)ofdatawerecreatedbySuper-powerhigh-techCorporationworldwide[16].
Letouzé,oneoftheBigDataforDevelopmentpioneers,hasdeveloped"the3Cs"ofBigData-presentinganotherperspective.
The3CsstandforBigData'Crumbs',BigData'Capacities'andBigData'Community';itfundamentallyframesBigDataasanecosystem,acomplexsystemactually,notasdatasources,setsorstreams.
Anditisbothinreferenceandoppositiontothe3VsofBigData[17].
Accordingtohisconcept,BigDataisnotjustdata—nomatterhowbigordifferentitisconsideredtobe;thisiswhyandwhereBigDataasafield—anecosystem.
GaryKing'sHarvardpresentationon"BigDataisnotaboutthedata"alsoandperhapshighlightsthatbigdataisfirstandforemost'about'theanalytics,thetoolsandmethodsthatareusedtoyieldinsights,turnthedataintoinformation,then,perhaps,knowledge[18].
The2nd'C'ofBigData,forCapacities,islargelyaboutthat—thetoolsandmethods,thehardwareandsoftwarerequirementsanddevelopments,andthehumanskills.
Thereis13aneedtobothconsideranddevelopcapacities,withoutwhichcrumbsareirrelevant.
Butit'snotjustaboutskillsandchips;it'salsoabouthowthewholequestionisframed.
Thisisofcourserelatedtotheconceptof'DataLiteracy',andtheneedtobecomesophisticatedusersandcommentators.
The3rdCofcommunityreferstothesetofactors—bothproducersandusersofthesecrumbsandcapacities;it'sreallythehumanelement—potentiallyit'sthewholeworld.
Figure3:The3C's"Diagram[17]Andtheresultingconcentriccircleswithcommunityasthelargersetareacomplexecosystem—withfeedbackloopsbetweenthem.
Forexamplenewtoolsandalgorithmsproducenewkindsofdata,whichmayinturnleadtothecreationofnewstartupsandcapacityneeds.
Letouzéandothers[17,18]arguethatthebasicpointisthatBigDataisnotbigdata;andthatquestionslike"howcannationalstatisticalofficeuseBigData"don'tmeanmuchorrathertheymissthepoint.
TherealimportantquestioniswhyandhowanNSO(NationalStatisticalOffice)shouldengagewithBigDataasanecosystem,partnerwithsomeofitsactors,becomeoneofitsactors,andhelpshapethefutureofthisecosystem,includingitsethical,legal,technicalandpoliticalframeworks.
ThisquestioncanthenbeexpandedtothesustainabledevelopmentactorsinterestedinusingbigdataandtobecomepartoftheBigDataEcosystem.
Thiswouldalsoinvolvetheroleofdevelopmentactorsasfacilitators,knowledgebrokersandconveningpowers.
14Thisreportisstructuredinasimilarway:fromanarrowfocusonBigDatatopromotingtheestablishmentofasystemsapproachtoBigData.
Thefocusofthisreportwillthusfocus(i)ontheactorsandtheirroleintheecosystem;(ii)thepotentialroleBigDatacanplayinthePolicyCycle,and(iii)StepstowardstheEcosystem'sapproachandUNESCAP'spotentialrole.
2.
2.
UseofBigDataThesheervolumeofdatagenerated,stored,andminedforinsightshasbecomeeconomicallyrelevanttobusinesses,government,andconsumers.
Inthecontextofpolicymaking,bigdatacanbeusedtoenhanceawareness(e.
g.
capturingpopulationsentiments),understanding(e.
g.
explainingchangesinfoodprices),and/orforecasting(e.
g.
predictinghumanmigrationpatterns).
Inmostcountries,publicsectorbodiesalsogatherenormousamountsofdatafromcensuses,taxreturns,andpublichealthsurveys,forexample.
Muchofthisdataistechnically"public,"butaccessingitisnotalwayseasy,andminingitforrelevantinsightscanrequiretechnicalexpertiseandtrainingthatorganizationsandgovernmentswithlimitedresourcescan'talwaysafford.
Makinggooduseofbigdatawillrequirecollaborationofvariousactorsincludingdatascientistsandpractitioners,leveragingtheirstrengthstounderstandthetechnicalpossibilitiesaswellasthecontextwithinwhichinsightscanbepracticallyimplemented.
Box2:TwitterExample:UseofMobileTechnologyforPerceptionAssessmentSince2010,Indonesiahaswitnessedsubstantialincreasesinfoodprices:thepriceofriceincreased51%betweenDecember2009andFebruary2012.
Withmorethan20millionTwitteruseraccountsinJakarta,awealthofdataisbeingproduceddaily.
PulseLabJakartaanalyzedTwitterconversationsdiscussingfoodpriceincreasesbetweenMarch2011andApril2013.
Taxonomies,thataregroupsofwordsandphraseswithrelatedmeanings,weredevelopedintheBahasaIndonesialanguagetoidentifyrelevantcontent.
Aclassificationalgorithmwastrainedtocategorizetheextractedtweetsaspositive,negative,confused,orneutraltoanalyzetheirsentiment.
Usingsimpletime-seriesanalysis,theresearchersquantifiedthecorrelationbetweenthevolumeoffood-relatedTwitterconversationsandofficialfoodinflationstatistics.
Arelationshipwasfoundbetweenretrospectiveofficialfoodinflationstatisticsandthenumberoftweetsspeakingaboutfoodpriceincreases.
Moreover,uponanalyzingfuelpricetweets,itwasfoundthatperceptionsoffoodandfuelpriceswererelated.
ThisbigdataexamplewascreatedbytheGlobalPulsetodemonstratetherelevancewithinthepolicycontexttoGovernmentofIndonesia.
[19].
ThepublicsectorcannotfullyexploitBigDatawithoutleadershipfromtheprivatesector[20].
TheconversationaroundDataPhilanthropy-atermwhichdescribesanewformofpartnershipinwhichprivatesectorcompaniessharedataforpublicbenefit-has15advancedsinceitsemergenceattheWorldEconomicForuminDavosin2011.
DiscussionsabouttheconceptofDataPhilanthropy,orprivatesectordatasharing,havegainedmomentumandmovedforward,reachingabroaderaudience.
Inanarticleabouttheissue,FastCompany'sCo.
Exist,summarized:'(t)henextmovementincharitablegivingandcorporatecitizenshipmaybeforcorporationsandgovernmentstodonatedata,whichcouldbeusedtohelptrackdiseases,averteconomiccrises,relievetrafficcongestion,andaiddevelopment.
Thepublicsectorisn't,however,theonlyonetogainfromDataPhilanthropy:companiesdonatingdatacangetadvantagefromittoo,especiallythosecompaniesinterestedinthesustainableeconomy.
Thesecompaniescouldenhancetheirroleincorporatesocialresponsibilitiesthusshapingtheirbranding.
Also,theirroleasstakeholdersmightchangeastheywillgettoinfluencepoliciesandpublicopinioninabroaderwaythanrelatedtotheirveryownbusiness[21].
Bigdataisshowingpromisetoimprove,andperhapssubstantivelychange,publicsectorandtheinternationaldevelopmentsectorinnovelways.
Ofgeneralinterestisthefactthatbigdataoftenisproducedatamuchmoredisaggregatedlevel,e.
g.
individualinsteadofacountrylevel.
Whereasaggregateddataglossesovertheoftenwide-rangingdisparitieswithinapopulation,disaggregateddataallowsdecisionmakersmoreobjectivelytoconsiderthoseportionsofthepopulationwhowerepreviouslyneglected.
2.
3.
BigDataandOpenDataInthecontextofpolicymaking,itisworthtoelaborateontheinterfacebetweenbigdataandthenewphenomenonof"opendata"-theyarecloselyrelatedbutarenotthesame.
Opendatabringsaperspectivethatcanmakebigdatamoreuseful,moredemocratic,andlessthreatening.
Whilebigdataisdefinedbysize,opendataisdefinedbyitsuse.
Butthosejudgmentsaresubjectiveanddependentontechnology:today'sbigdatamaynotseemsobiginafewyearswhendataanalysisandcomputingtechnologyimprove.
Alldefinitionsofopendataincludetwobasicfeatures:thedatamustbepubliclyavailableforanyonetouse,anditmustbelicensedinawaythatallowsforitsreuse.
Opendatashouldalsoberelativelyeasytouse,althoughtherearegradationsof"openness".
16Figure3:TheInterfaceofBigDataandOpenData[22]ThediagraminFigure3mapstherelationshipbetweenbigdataandopendata,andhowtheyrelatetothebroadconceptofopengovernment.
Thereareafewimportantpointstonote:a)Bigdatathatisnotopenisnotdemocratic:Sectiononeofthediagramincludesallkindsofbigdatathatiskeptfromthepublic–likethedatathatlargeretailersholdontheircustomers,ornationalsecuritydata.
Thiskindofbigdatagivesanadvantagetothepeoplewhocontrolit.
b)Opendatadoesnothavetobebigdatatomatter:Modestamountsofdata,asshowninsectionfour,canhaveabigimpactwhenitismadepublic.
Datafromlocalgovernments,forexample,canhelpcitizensparticipateinlocalbudgeting,choosehealthcare,analyzethequalityoflocalservices,orbuildappsthathelppeoplenavigatepublictransport.
c)Big,opendatadoesn'thavetocomefromgovernment:Thisisshowninsectionthree.
Moreandmorescientistsaresharingtheirresearchinanew,collaborativeresearchmodel.
Otherresearchersareusingbigdatacollectedfromsocialmedia–mostofwhichisopentothepublic–toanalyzepublicopinionandmarkettrends.
But,whengovernmentsturnbigdataintoopendata,itisespeciallypowerful:Governmentagencieshavethecapacityandfundstogatherverylargeamountsofdata,andopeningupthosedatasetscanhavemajorsocialandeconomicbenefits.
Bothbigdataandopendatacantransformbusiness,government,andsociety–andacombinationofthetwoisespeciallypotent.
Bigdatagivesunprecedentedpowertounderstand,analyze,andultimatelychangetheworldwelivein.
Opendataensuresthatpowerwillbesharedbearinghugepotentialtotransformthewaypoliciesaremade.
173.
TheBigDataEcosystemUnlikeinotherareas,thestakeholdersintheBigDataspherearenotyetwellconnectedandsomeprocessesneedtobeinplacetobringthemtogether.
Makinggooduseofbigdatawillrequirecollaborationofvariousactorsincludingdatascientistsandpractitioners,leveragingtheirstrengthstounderstandthetechnicalpossibilitiesaswellasthecontextwithinwhichinsightscanbepracticallyimplemented[22].
Policystakeholdersactattheinternational,regional,nationalandlocallevel.
Whenlookingatthegovernmentactors,nosingletypeofresponsibleauthorityemergesasaclearleaderintheimplementationofinnovativedataforpolicyinitiatives,withtheclearimplicationthatthereareopportunitiesformanydifferentstakeholdersandactors.
3.
1.
TypesofStakeholdersTheEUdataforpolicyreport[23]distinguishesbetweenthefollowingtypesofstakeholders:globalandEuropeanpolicymakers;nationalpolicymakers;regionalpolicymakers;statisticaloffices;scienceandR&Dorganisations;databrokers;privateprovidersofdataanalyticsandvisualisationtools;civilsocietyandthepolicyanalysis/evaluationcommunity.
Forthepurposeofoutliningtherelevantstakeholders,thisreportadoptstheEUstakeholdercategories.
IntheEUforexample,BigDataisstimulatedtopromotejobsandeconomicgrowth,topromoteindustrialleadershipandanopensociety(opendata).
ItisconnectedtothemanysocietalchallengesthattheEuropeanCommissionhasdefined,amongwhichare'health,demographicchangeandwellbeing','smart,greenandintegratedtransport'and'climateaction,environment,resourceefficiencyandrawmaterials'.
However,noprojectscouldbefoundinwhichtheEuropeanCommissionusesbigdataitselffordirectuseinitsownpolicycycle.
Onanationalpolicy-makingscale,bigdataisoftenusedintheareasoftransport,whereinnovativesensor-dataprovidesrelevantinformation.
Moreover,itisusefulindetectingfraud,reducingcrimeandimprovingnationalsecurity,bothviadefenceandintelligence.
Nationalpolicymakerspotentiallypossessalotofdatathatcouldbeusedforinformedpolicymakingusingbigdataanalyses.
Openingupthesedatacouldbeafirststep(opendata).
Furthermore,theorganisationsofthesepolicymakershavesignificantfinancialmeanstosetupprojectsandimprovebigdataforpolicy.
Attheregionallevelbigdatacouldaddresspolicyissuesconcerningtraffic,roadsafety,criticalinfrastructure,wastemanagement,safetyandsecurityandpublichealth.
Incontrasttonationalpolicymaking,dataforregionalpolicyfocusesmoreonthepolicyimplementationinsteadofagendasetting.
Thestatisticalofficesusebigdatatoacquirebetterofficialstatisticsforpolicymeans.
Thesemayconcernallsortsofpolicyareas.
Societalchallengesthatcouldbeaddressedare,forexample,energyefficiency,infrastructure,smarttransportanddemographicchange.
Themostrelevantresourcesthatthestatisticalofficeshaveareknowledgeand18skillsrelatedtostatisticallyanalysinglargesetsofdata.
Theymayalsohavetheneededtechnologicalinfrastructuretostoreandprocessbigdata.
Theyhavefinancialmeanstoacquireandanalysedataforofficialpolicy.
Still,theymayhavetoexpandtheexperienceandITknowledgeandequipmentneededforbigdata.
Mostpilotsareperformedincooperationwithexternalinstitutes.
Themainbenefitsofbigdataforthestatisticalofficesisimprovingtheaccuracy,timelinessandrelevanceoftheirstatisticsandreducingcosts.
Forexample,usingsocialmediadataandhavingaccesstodataaboutofflineandofflineretailrevenuesislessexpensivethanlarge-scalesurveys(re-usingandmatchingdataversuscollectingdata).
Thesciencecommunitysupportspolicymakersinallpolicyareas,onallgovernmentallevelsandinallstepsofthepolicycycles.
Concerningsciencepolicy,themainpolicyquestionshaveincludedhowtopromoteanenvironment,whichprotectsintellectualpropertyandsupportsthemosteffectiveorganizationofdisciplinesandteamsandresources.
Steeringthelargeresourcesdevotedtoresearchintothemostusefulandbeneficialchannelscanbeofgreatbenefittosociety,andthisareahasbeenonewherethereisgreatsophisticationintheanalysisandmuchdataavailable.
Thesciencecommunityhasknowledgeandskillsrelatedtostatisticallyanalysinglargesetsofdataandusinganevidence-basedapproachwhenresearchingthedata-drivenapproaches.
Theyoftenalsohavetheneededtechnologicalinfrastructuretostoreandprocessbigdata.
Theyhavefinancialmeanstoconductresearch.
Moreovertheyhavethepossibilitytoconnectmultipledisciplinesintheirresearch(asthedatacentresdemonstrate).
Lastly,theypossessorhaveaccesstoavastamountoflargedatasets(e.
g.
climatedata,civilengineeringdata,socialandbehaviouraldata)andcanthusmoreeasilyconnectdifferentdatasets.
Databrokerscouldprovidetheirdataforallkindsofsocietalchallengesand/orpolicyareas.
Thoseareusuallycompaniesthatcollectinformation,includingpersonalinformationaboutusers,fromawidevarietyofsourcesforthepurposeofresellingsuchinformation.
Anexampleishealthcare,inwhichGoogleFlueTrendsisactive.
Databrokersoftendonotanalyseoractuallyusethedata;theyoftenonlyprovideitfortheotheractors.
Databrokershaveastheirmainresourcedatasetsonspecificgroupsoronsocietiesaslarge.
Furthermore,theyhaveknowledgeofandskillsindatacollectionandanalysis,forwhichtheyhavededicatedtools.
Asmostofthedataiscommerciallytraded,theyhavethefinancialmeansandincentivestoinvestintheimprovementofdatacollection,storageandanalysis.
3.
2.
RolesofStakeholdersintheEcosystem19Table1:RolesofStakeholdersinDataEcosystemGovernmentsMulti-NationalOrganizationsStatisticalBodiesR&DBodiesCivilSocietyPrivateProvidersDataxxxxxFinancialResourcesxxxxxStandardsandRegulatoryFrameworksxxSkillsandKnowledgex(x)xxBrokering,Facilitation,CapacityStrengtheningxxxxITInfrastructurexxxGovernmentsshouldempowerpublicinstitutionstorespondtothedatarevolutionandputinplaceregulatoryframeworksthatensurerobustdataprivacyanddataprotection,andpromotethereleaseofdataasopendatabydataproducers,andstrengthencapacityforcontinuousdatainnovation.
Multinationalorganizations,donors,governmentsandsemi-publicinstitutionsshouldinvestindata,providingresourcestocountriesandregionswherestatisticalandtechnicalcapacityisweak.
Theyshoulddevelopinfrastructuresandimplementstandardstocontinuouslyimproveandmaintaindataqualityandusability;keepdataopenanduseablebyall.
Theyshouldalsofinanceanalyticalresearchinforward-lookingandexperimentalsubjects.
Internationalandregionalorganizationsshouldworkwithotherstakeholderstosetandenforcecommonstandardsfordatacollection,production,anonymization,sharingandusetoensurethatnewdataflowsaresafelyandethicallytransformedintoglobalpublicgoods,andmaintainasystemofqualitycontrolandauditforallsystemsandalldataproducersandusers.
Theyalsoshouldsupportcountriesintheircapacity-buildingefforts.
Statisticalsystemsshouldbeempowered,resourcedandindependent,toquicklyadapttothenewworldofdatatocollect,process,disseminateandusehigh-quality,open,disaggregatedandgeo-codeddata,bothquantitativeandqualitative.
Allpublic,privateandcivilsocietydataproducersshouldsharedataandthemethodsusedtoprocessthem,accordingtoglobally,regionally,ornationallybrokeredagreementsandnorms.
Theyshouldpublishdata,geospatial20informationandstatisticsinopenformatsandwithopentermsofuse,followingglobalcommonprinciplesandtechnicalstandards,tomaintainqualityandopennessandprotectprivacy.
Governments,civilsociety,academiaandthephilanthropicsectorshouldworktogethertoraiseawarenessofpubliclyavailabledata,tostrengthenthedataandstatisticalliteracy("numeracy")ofcitizens,themedia,andother"infomediaries",ensuringthatallpeoplehavecapacitytoinputintoandevaluatethequalityofdataandusethemfortheirowndecisions,aswellastofullyparticipateininitiativestofostercitizenshipintheinformationage.
Theprivatesectorshouldreportonitsactivitiesusingcommonglobalstandardsforintegratingdataonitseconomic,environmentalandhuman-rightsactivitiesandimpacts,buildingonandstrengtheningthecollaborationalreadyestablishedamonginstitutionsthatsetstandardsforbusinessreporting.
Civilsocietyorganizationsandindividualsshouldholdgovernmentsandcompaniesaccountableusingevidenceontheimpactoftheiractions,providefeedbacktodataproducers,developdataliteracyandhelpcommunitiesandindividualstogenerateandusedata,toensureaccountabilityandmakebetterdecisionsforthemselves.
Academicsandscientistsshouldcarryoutanalysesbasedondatacomingfrommultiplesourcesprovidinglong-termperspectives,knowledgeanddataresourcestoguidesustainabledevelopmentatglobal,regional,national,andlocalscales.
Theyshouldmakedemographicandscientificdataasopenaspossibleforpublicandprivateuseinsustainabledevelopment;providefeedbackandindependentadviceandexpertisetosupportaccountabilityandmoreeffectivedecision-making,andprovideleadershipineducation,outreach,andcapacitybuildingefforts.
Therefore,thedifferentstakeholdersforbigdata,whichincludesownersandusers,shouldideallyemergeintoa"globaldatasystem",orbigdataecosystem,tosupportpolicymaking.
However,thechallengewillbeinhowtobringthesedifferentstakeholdersandsystemstogethertomakethedatarevolutionhappen.
Thesestakeholdersareoperatingwithintheirsystemsandproceduresanditisimportantthatforaandplatformsarebeingestablishedandmanagedeffectivelytomakethebigdatasystemwork.
EffectiveapplicationofBigDataforDevelopmentwouldalsorequirechangesinthedecision-makingprocess,whichcustomarilyreliesontraditionalstatistics.
GiventhehighfrequencyofBigData,amoreresponsivemechanismwillneedtobeputinplacethatallowsthegovernmenttoprocesstheinformationandactquicklyinresponse.
Also,sinceBigDataisoftenunstructuredandrelativelyimprecise(comparedtoofficialstatistics),governmentofficialsalsohavetolearnhowtoeffectivelyinterpretandmakeuseoftheinformationprovidedbyBigData.
Thisrequirescapacitybuildingtoturndecisionmakersintomoresophisticateddatausers.
214.
BigDataandPolicyMakingBigdatastrategiesfordevelopmentcanbeimportanttoolstoformulatepoliciesthatalsohelpsuccessfullyimplementingtheSDGs.
However,manyemergingeconomiesordevelopingcountriesarestillstrugglingwithcollectingandmanagingmuchsmallerdatasetsandstatistics.
Whilealotof"smaller"dataexists[24],itisoftennotintegrated,patchyandoflowquality.
Also,thesestatisticsareoftentop-downandaremissingafeedbacklooptocommunities.
Thebigdatadiscussionmightoverlooktheveryfactthatcapacityconstraintsareonechallengethatneedstobesystematicallyaddressedaspartofthebigdatadiscussion.
4.
1.
BestPracticesThediscussionofdata-drivenapproachestosupportpolicymakingcommonlydistinguishesbetweentwomaintypesandusesofdata.
Thefirstistheuseofpublicdatasets(administrative(open)dataandstatisticsaboutpopulations,economicindicators,educationetc.
)thattypicallycontaindescriptivestatistics,whicharenowusedonalargerscale,usedmoreintensively,andlinked.
Thesecondisdatafromsocialmedia,sensorsandmobilephones,whicharetypicallynewsourcesforpolicymaking.
Bestpracticesarestillevolvingwhereinnovativeapproachescomplementexistingusesofdataforpolicy.
AccordingtoastudyfortheEU,themostcommonusesofbigdatainpolicymakingincludepilotswherenewsourcesofdataarebeingusedforagenda-settingandpolicyimplementation;useofopendatafortransparency,accountabilityandparticipationandusingadministrativeandstatisticaldataformonitoringtheoutputsandimpactofpolicies.
Below(Box4)anexampleofastateofthearttool(APPA)thatisrevolutionizingelementsofpolicymaking.
CountriesintheAsiaandthePacificregion,includingamongothersSingapore,Indonesia,RepublicofKorea,andthePhilippines,aswellastheUSandJapanarealreadysuccessfullyinnovatingwithandopeningupdatatosolvecomplexpolicyproblems,increaseallocativeefficiencyandimprovedemocraticprocesses[25].
DataanalysisintheprocesscomponentofthePolicyCircleismorecomplexthaninproblemidentificationbecausepolicymakersweightheirdecisionsonanumberofcriteria.
Dataanalysisexpandsfromthetechnicalaspectsofanissueandfocusesonthepoliticalcostsandbenefitsofpolicyreform[3]topositthatpolicymakerstendtomaketheirdecisionsbasedonanumberofcriteria,including:1)thetechnicalmeritsoftheissue;2)thepotentialaffectsofthepolicyonpoliticalrelationshipswithinthebureaucracyandbetweengroupsingovernmentandtheirbeneficiaries;3)thepotentialimpactofthepolicychangeontheregime'sstabilityandsupport;4)theperceivedseverityoftheproblemandwhetherornotthegovernmentisincrisis;and5)pressure,support,oroppositionfrominternationalaidagencies[26].
22Rather,bigdataisanadditionalmeansthathashugepotentialtoimprovepolicies.
Interestingly,anEUstudy[27]findsthatmostlybigdataisusedattheearlystageofthepolicycycle,bymakinguseofdataandforesight,agendasetting,problemanalysisandforidentificationanddesignofpolicyoptions.
Accordingtothestudy,lessthanathirdofinitiativeshaveafocusonthemiddle-stagepolicycyclesfortheimplementationofpoliciesandinterimevaluation.
Also,thisstocktakefindsthatbigdataisnot(yet)playingacrucialroleinpolicymaking.
Ifatall,itisusedattheagendasettingstageand/orevaluationstageofpolicymaking.
Oneofthereasonsmightbethatbecausetheecosystemisnotyetfunctioningandcrucialelements,suchasstandardsandframeworksarestillmissing.
Nationalgovernmentsandotherpolicymakersarejuststartingtosystematicallyengagewithbigdataforpolicymaking.
4.
2ThePolicyCycleThereareopportunitiesforfull-scaleimplementationofdata-drivenapproachesacrossallstagesofthepolicycycle,includingevaluationandimpactassessment.
Thefollowingsectionidentifiessomedatadrivenapproachesineachstepofthepolicycycle:Figure5:Policy-MakingProcess[28]23PolicyCycleStep1-AgendaSetting:Theagendasettingstageisoneofthemajorstepsinthepolicymakingcycle.
Onceaproblemrequiringapolicysolutionhasbeenidentified,theprocessofpolicydevelopmentincludeshowtheproblemisframedbyvariousstakeholders(issuesframing),whichproblemsmakeitontothepolicymakingagenda,andhowthepolicy(orlaw)isformulated.
Together,thesesteps,determinewhetheraproblemorpolicyproposalisactedon.
Activitiesinpolicydevelopmentincludeadvocacyandpolicydialoguebystakeholdersanddataanalysistosupporteachstepoftheprocess.
Issueframinginfluencesstakeholders'abilityofgettingtheissueonthepolicymakers'agendasothataproblemisrecognizedandpolicyresponseisdebated.
Issueframingoftensetsthetermsforpolicydebate.
Agendasettingreferstoactuallygettingthe"problem"ontheformalpolicyagendaofissuestobeaddressedbypresidents,cabinetmembers,Parliament,Congress,orministersofhealth,finance,education,orotherrelevantministries.
Stakeholdersoutsideofgovernmentcansuggestissuestobeaddressedbypolicymakers,butgovernmentpolicymakersmustbecomeengagedintheprocessforaproblemtobeformallyaddressedthroughpolicy.
Governmentpolicymakingbodies"canonlydosomuchinitsavailabletimeperiod,suchasthecalendarday,thetermofoffice,orthelegislativesession.
Theitems,whichmakeittotheagendapassthroughacompetitiveselectionprocess,andnotallproblemswillbeaddressed.
Inevitably,somewillbeneglected,whichmeansthatsomeconstituencywillbedenied.
Amongthepotentialagendaitemsareholdoversfromthelasttimeperiodorareexaminationofpoliciesalreadyimplementedwhichmaybefailing"[29].
Atanygiventime,policymakersarepayingseriousattentiontorelativelyfewofallpossibleissuesorproblemsfacingthemasnationalorsubnationalpolicymakers.
Indecentralizedsystems,sometimesissuesareplacedontheagendaofvariouslevelsofgovernmentsimultaneouslytocoordinatepolicymaking.
Inordernottomakethingsoverwhelming,itiskeytobeginwithquestionsthatneedtobeansweredinthepolicymakingprocess,notwithdata.
Oncethesettingfortheanalysisisdefined,thefocusoftheresearchcanmovetothebehaviorsofinterestandtheconsequentdatagenerationprocess.
Keyexemplarystrategiesdescribedintheboxespotentiallycanmovethepolicyarenaforwardinaproductiveway.
Theyarebynomeansexhaustive.
Also,literatureisactuallymissingonhowexactlybigdatahasinfluencedpolicymakingvs.
traditionaldata.
PolicyCycleStep2-PolicyFormulation:Policyformulationisthepartoftheprocessbywhichproposedactionsarearticulated,debated,anddraftedintolanguageforalaworpolicy.
Writtenpoliciesandlawsgothroughmanydraftsbeforetheyarefinal.
Wordingthatisnotacceptabletopolicymakerskeytopassinglawsorpoliciesisrevised.
Policyformulationincludessettinggoalsandoutcomesofthepolicyorpolicies[30].
Thegoalsandobjectivesmaybegeneralornarrowbutshouldarticulatetherelevantactivitiesandindicatorsbywhichtheywillbeachievedandmeasured.
Thegoalsofapolicycouldinclude,forexample,thecreationofgreateremploymentopportunities,improved24healthstatus,orincreasedaccesstoreproductivehealthservices.
PolicyoutcomescouldincludeforexampleensuringaccesstoARVtreatmentforHIVintheworkplaceoraccesstoemergencyobstetriccareforpregnantwomen.
Goalsandoutcomescanbeassessedthroughanumberoflenses,includinggenderandequityconsiderations.
ActivitiesRelatedtotheProcess—Advocacy,PolicyDialogue,andDataAnalysis.
Whileissuesframing,agendasetting,andpolicyformulationarestagesthatpoliciesgothrough,eachofthesestagescanincludeanumberofactivities,namelyadvocacy,policydialogue,andanalysisofevidencerelatedtotheproblemandpolicyresponses.
Theinterpretationofthisinformationwillincludevariouspolicystakeholders-theseincludethelegislature,CSO'sandotherrelevantstakeholders.
Theexecutivewillhavetoproduceactionableinsightswiththepossibleobjectiveofinfluencingthebehaviorsofinterestconsidered.
Thisalsoincludesmappingthelandscape–understandingthepolicyarena'sissuesandcurrentchallenges.
Keyplayersandstakeholdersinthepolicyarenaandtheirrelationshipstoeachotherneedtobeidentifiedandmobilized.
Bigdatanowallowcreatingmultiplescenariostounderstandhowthepolicylandscapemayevolve.
Also,communityparticipationcanbeenhancedwithmobiletechnology.
PolicyCycleStep3-PolicyAdoption:Thepolicyadoptionprocessistypicallystillapplyingtheconventionalpolicyinstitutionalizationmethods-draftinglawsandregulations.
However,thedisseminationofnewpoliciescanbefasterandwiderwiththeInternet,appsetc.
Thepotentialtothecomplianceandtake-upofnewpoliciescanincreasedramatically.
Ofcourse,allthisinformationisuselessunlessitisusedtogenerateinsightsthatleaderscanacton.
Fortunately,advancesinanalysisandvisualisationtools(interactivecharts,infographics,deepzoomingapplications,etc.
)meanitisnowfeasibletobringgranularandup-to-dateevidencetobearonleadershipchallenges.
Thisappliesacrosstheboard–fromanalysingandoptimisingtheimpactofpolicies,throughtogatheringandactingonfeedbackfromcitizensoncertainpolicies.
Inmanyinstances,importantsourcesofbigdataforlearningliveoutsidetraditionalorganisationalboundaries[31].
PolicyCycleStep4-PolicyImplementation:Procedures,guidelinesandresourcesneedtobemadeavailableforpolicyimplementation.
SIMGovernment(Box3)isoneofthefewexamplesavailablewherebigdataisusedforpolicyimplementation.
Box3:SIMGovernmentLikethepopularcomputergameSimCity,APPAcreatesaSimGovernmentforpolicymakerstobuildpossiblepoliciesandthentesttheeffectsofthosepoliciesinarealisticenvironment.
Astheamountofdatagrowsandtheanalytictechniquesbecomemoresophisticated,itispossibletomeasuretheimpactofpoliciesonotherissuelandscapes.
Forexample,policymakerscouldmodelhowanewhealthpolicywillaffectenvironmentalandeducationalissues,alongwithhealthissues.
25AmajoradvantageofAPPAisthatitwillalsohelpinidentifyingtheundesiredresultsofpolicies.
Withcurrentpolicymaking,ittakestimetocollectthedataandobservetheresultsofapolicy.
Thisdelayoftenworsenstheundesiredeffectsofapolicy-sometimesforyears.
WithAPPA,policymakerscanspotandpreventtheundesiredeffectsoftheirpoliciesbeforeimplementation.
[32]PolicyCycleStep5-PolicyEvaluation:Policiescanbeevaluatedinavarietyofinformalandformalmethodsandthiscanbeinitiatedanddrivenbyawholerangeofdifferentactors,suchasthelegislature,CSO's,theexecutive,academiaorotherrelevantstakeholders.
However,formalmethodstendtobedifficulttocarryoutandinformalmethodscanberiddledwithbias.
Policiescanbeevaluatedwhiletheyarebeingimplementedoraftertheyhavebeenimplemented.
Theyaredifficulttoevaluatewhentheyaimtoaccomplishbroadconceptualgoals,havecompetingobjectives,orpossessmultipleobjectives.
Mostpoliciesfailtobeevaluatedduetoassessmentdifficultiesandthetendencyofthepolicyprocesstofavorthestatusquo.
Also,policyevaluationscanbeexpensivetodo.
Publicadministrationsarenotnecessarilywellequippedtodesignevaluations-scope,sequencing,etc.
Externalpartiesmightbeavailable,butthischoicehasnotbeenwidelyapplied[33].
Inthepolicyevaluationandpolicyrevisionselements,bigdatacanpotentiallyplayabigroleasitcanprovidefeedbackloopsandinformationthatwaspreviouslynotavailable.
Box4:AgilePredictivePolicyAnalysis(APPA)Agilepredictivepolicyanalysis(APPA)isbuiltupontheconceptsbehindotherdata-basedpolicymakingfunctionssuchastheObamaadministration'sPortfolioStatIT[34].
Dataisblendedfromvarioussourcestocreateadashboardthatdisplayskeyperformanceindicatorswheredecisionmakerscancreateandmonitorpolicies.
Thisgivespolicymakersnear-real-timefeedbackontheperformanceofpoliciesandgovernancedecisions.
ThegoalofAPPAistonotonlyusedatatoaccuratelyreportonthecurrentstateoftheagenciesandpoliciesbuttocreateaccuratemodelsofthelandscape,protagonists,andthepolicystruggletocreatethemostlikelyscenarios.
Thisisaccomplishedbyusingtransactionaldatasourcesfromagencyoperationsandusingdatasciencetechniquessuchasmachinelearningandpredictiveanalyticstobettermodelagencydecisions.
Therelationshipsbetweentheagenciesandotherpolicystakeholdersaremodeledalongwithanyrelevantenvironmentalfactorsinthepolicylandscape.
Thisallgoestowardcreatingasimulationofthepolicylandscape,whichgivesboththecurrentstatusandfuturescenarios.
26Thereisnothingnewinusingresearchtechniquesdevelopedinacademiatoanalyzedatabypublicpolicypractitioners.
Onecanseeacyclewherepublicagenciescreatethedataandanalyticalchallengesthatleadtoacademicresearchinmoreeffectivepolicymakingtechniques,whichinturnleadstoevenmoredatacollectionandmorecomplexanalyticalchallenges.
APPAisthelatestiterationinthiscyclewherecomplexitytheoryanddatasciencewillleadtomoresophisticatedpolicymaking,whichanticipatespolicyeventsratherthanjustreactstothem.
Bigdatatechnologiesalonearenot,however,asilverbulletfortransformingthepublicsector.
Underlyingdataissueslikequality,standardsandbiasstillneedtoberecognizedandaddressed.
Andgovernmentsmusthavethecapabilitytoconduct,interpretandconsumetheoutputsofdataandanalyticsworkintelligently.
Thisisonlypartlyaboutcutting-edgedatascienceskills.
Justasimportant-ifnotmoreso-isensuringthatpublicsectorleadersandpolicymakersareliterateinthescientificmethodandconfidentcombiningbigdatawithsoundjudgment.
Governmentswillalsoneedthecouragetopursuethisagendawithstrongethicsandintegrity.
Thesametechnologythatholdssomuchpotentialalsomakesitpossibletoputintensepressureoncivilliberties.
5.
ChallengesandOpportunitiesSeveralchallengesandconsiderationswithbigdatamustbekeptinmind.
Thisreporttouchesonsomeofthemanddoesnotpretendtoprovideanswersandsolutionsbutrathertopromotediscussion.
AWorldBankstudy[35]showsthatabouthalfofthe155countrieslackadequatedatatomonitorpovertyand,asaresult,thepoorestpeopleinthesecountriesoftenremaininvisible.
Duringthe10-yearperiodbetween2002and2011,asmanyas57countries(37percent)hadnoneoronlyonepovertyrateestimate.
Lackofwell-functioningcivilregistrationsystemswithnationalcoveragealsoresultsinseriousdatagaps.
5.
1.
ChallengesInstitutionalFrameworksInstitutionalframeworks,meaningtheinstitutionsthatarerequiredtoprotectpillarsofdemocracysuchasprivacy,areoftennotinplacewhenitcomestobigdata.
Thisisakeychallengethatneedstobeaddressedinordertoscaleuptheuseanduseabilityofbigdataforsustainabledevelopment.
Privacy,definedastherightofindividualstocontrolwhatinformationrelatedtothemmaybedisclosed,isapillarofdemocracy,andprotectionsmustbeputinplacetoavoidcompromisingthisbasichumanrightinthedigitalage.
PrivacyisanoverarchingconcernforanyonewishingtoexploreBigDatafor27development,sinceithasimplicationsforallareasofwork,fromdataacquisitionandstoragetoretention,useandpresentation.
Inmanycases,theproductionofdataitselfraisesconcerns,aspeoplemaybeunawareofthesheerquantityortypesofdatatheyaregeneratingonadailybasis,aswellasthatdatatheyunknowinglyconsenttothecollectionandusageofwithoutunderstandinghowitmaybeused[36].
Inthiscontext,itisimportanttonotethatsuitablelegalframeworks,ethicalguidelinesandtechnologicalsolutionsforprotecteddatasharingareatthecenterofeffortstoleverageBigDatafordevelopment.
DigitalDivideAlthoughthedatarevolutionisunfoldingaroundtheworldindifferentwaysandatdifferentspeeds,thedigitaldivideisclosingfasterthanmanyhadanticipated.
Theavailabilityandtypesofdigitaldata,however,differfromcountrytocountry.
Forinstance,countrieswithhighmobilephoneandInternetpenetrationrateswillproducemoredatadirectlygeneratedbycitizens,whilenationswithlargeaidcommunitieswillproducemoreprogram-relateddata.
Dataalsovariesbetweenagegroups,economicincomebrackets,genderandgeographiclocation.
ThesetypesofbiasesmustbeaddressedinthewayBigDatacaninfluencepolicies,andparticularattentionmustbegiventothecountriesthatareproducinglessdataand/orhavelesscapacityindataanalyticstoavoidaddingnewfacetstodigitaldivide[37].
Itisimportanttoalsohighlightthatitisnotonlyadigitaldividebetweencountries,butalsowithincountries.
Arethepoorestofthepoorabletoaccessanyofthetechnologiesorservicesthatwouldcollecttheirdata-aretheythenrepresentedinbigdatastatisticsandfiguresAnalysisofbigdataresultshastotakethisintoconsideration.
AccessandPartnershipsAlthoughmuchofthepubliclyavailableonlinedatahaspotentialutilityfordevelopmentpurposes,privatesectorcorporationsholdagreatdealmoredatathatisvaluablefordevelopment.
Companiesmaybereluctanttosharedataduetoconcernsaboutcompetitivenessandtheircustomers'privacy.
Workingwithbigdatarequiresanewformofpartnershipbetweendatamakers,datausers(seedatasystemsectionabove),anddatastoragestakeholders/institutionstoensurethatthepotentialofbigdataisrealized.
Itisanewwayofworking,andthechallengeofbridgingtheworldstogetherisabigone.
AnalyticalandCapacityChallengesTheprocessofminingBigData(usingBigDataanalyticstechniquestoextractrelevantinformation)containscertainanalyticalrisksthatmayreducetheaccuracyoftheresults.
AnalyzingBigDataforpolicyinputsandevaluationsposesdifferentchallengesthatareinpartmethodological,orrelatedtointerpretationaccuracy,methodsofanalysis,anddetectionofanomalies[38],whichwillbenotfurther,elaboratedinthisreport.
28Thecapacitytoeffectivelyutilizeallthepotentialthatbigdatabringsalongisstillverylimited.
Theinstitutionalframeworksmissingalsoimpedethestrengtheningofcapacitiesofdifferentstakeholdersandtheirroles.
5.
2.
OpportunitiesTheuseforbigdataforpolicymakingisaboutturningimperfect,complexandoftenunstructureddataintoactionableinformation.
Despitethemanychallengesthatbigdataanalysispresents,understandingthegrowingamountofdigitalinformationhumancommunities'producecanbeinvaluableinprovidingthemwithsupportandprotection.
Citizen-FocusandParticipationBigdataoffersachanceforpolicy-makingandimplementationtobemorecitizen-focused,takingaccountofcitizens'needs,preferencesandactualexperienceofpublicservices,asrecordedonsocialmediaandotherplatforms[39].
AscitizensexpresspolicyopinionsonsocialnetworkingsitessuchasTwitterandFacebookorrateorrankservicesoragenciesongovernmentapplications,policymakersalsohaveaccesstoahugerangeofdataoncitizens'actualbehavior,asrecordeddigitallywhenevercitizensinteractwithgovernmentadministrationorundertakesomeactofcivicengagement,suchassigningapetition.
Dataminedfromsocialmediaoradministrativeoperationsinthiswayalsoprovidearangeofnewdatawhichcanenablegovernmentagenciestomonitor–andimprove–theirownperformance,forexamplethroughlogusagedataoftheirownelectronicpresenceortransactionsrecordedoninternalinformationsystems,whichareincreasinglyinterlinked.
Andtheycanusedatafromsocialmediaforself-improvement,byunderstandingwhatpeoplearesayingaboutgovernment,andwhichpolicies,servicesorprovidersareattractingnegativeopinionsandcomplaints,enablingidentificationofafailingschool,hospitalorcontractor,forexample.
Theycansolicitsuchdataviatheirownsites,orthoseofsocialenterprises.
Andtheycanfindoutwhatpeopleareconcernedaboutorlookingfor.
Efficientprocedurestodrawlinksbetweenlarge-scaledata-processingtechnologiesandexistingexpertknowledgeinmajorpolicydomainswouldpotentiallyofferchancestomakepolicydevelopmentprocessesmorecitizen-focused,takingintoaccountpublicneedsandpreferencessupportedwithactualexperiencesofpublicservices.
Bigdatacancontributetothetransformationofcitizen-staterelations.
Datacanbeusedtotrackserviceprovision,enablecitizenstoreallocatelocalbudgets,makechangesintheircommunities,holdtheirgovernmentstoaccountandtoparticipatebetterindemocraticprocessestoensuretheirneedsandconcernscount-oftenforthefirsttime.
Evidence/MoreandBetterAnalyticsThenotionthatpolicydecisionshouldbebasedonsoundevidencehasbecomewidelyadoptedbymanypublicadministrations.
Strengtheningscience-policyinterfaceisalso29highlightedintheRio+20outcomedocument"TheFutureWeWant[40]aswellasthe2030AgendaforSustainabledevelopment.
Datatechnologiesareamongstthevaluabletoolsthatpolicymakershaveathandforinformingthepolicyprocess,fromidentifyingissues,todesigningtheirinterventionandmonitoringresults.
Moredataoftenmeanswecandomorewithanalytics,especiallyadvancedanalytics.
Bigdataandnewformsofdatacollectionwillgivecitizensnewinformationtheyneedtolivebetterlivesandearnmoresecurelivelihoods.
Theycantellpeoplethebesttimetoavoidtraffic,whenbesttoplantcrops,andwhichwaterholesarefreefromarsenic,fluoride,iron,andparasites.
VarietyValidationisakeysuccessfactortobenefitfromanalyticalinsights.
Thevarietyofdataavailablenowadaysmakesiteasiertodetermineifcertaininsightsareconsistentwithdatafrommultiplesources(triangulation).
Giventhelowcostofattainingandthesizeofavailabledata,replicationisnowofteneasier,andanythingonlinecanbeeasilytested.
Real-TimeInformationReal-timelinessreferstodatabeingavailablemuchfasterandsometimesinreal-time.
Internaldatacanbeavailableinaweek;clickstreamdatacouldprobablybeobtainedanhourafteritiscaptured-providedtheinitialsetupandcodinghasbeendone—andsocialmediacommentscanbewatchedinreal-time.
Itiswidelybelievedthattheuseofinformationtechnologycanreducethecostofpublicserviceswhileimprovingitsquality.
Datacanberoutinelycapturedandcreatedintheday-to-daybusinessofgovernment.
Itisimportanttonotethat,forthepurposeofglobaldevelopment,"realtime"doesnotalwaysmeanoccurringimmediately,butratherreferstoinformationthatisproducedandmadeavailableinarelativelyshortandrelevantperiodoftimeandwithinatimeframethatallowsactiontobetakeninresponse,creatingafeedbackloop.
EarlyWarningSystemDatacollectedthroughnewtechnologiescanactasanearly-warningsystem.
Evenifwedonotknowatthemacroleveltheprecisenumberofclinicsorpharmaciesthatstockvitalmedicines,ifpeoplecanalerttheirgovernmentviaSMStostockouts,thissignalsproblemsinacertainarea,meaningthatactioncanbetakenbeforeafulldatasetisavailable.
EconomicValueGoodqualitydatayieldnotonlysocialbenefits,butalsorealeconomicreturns,suchthat,inthemediumterm,adatarevolutioncouldpayforitself.
First,ifgovernmentsinvestinbettereconomicdata,thiscanimproveinvestorconfidence.
TheIMFhasfoundthat,ifcountriesinvestinbetter-qualitydata,itischeaperforthemtoborrow30internationally.
Itinvestigatedtheeffectofitsdatastandardsonsovereignborrowingcostsin26emergingmarketanddevelopingcountriesandestimatedthatcountriesthatsignuptoitsmorestringentdatastandardreduceborrowingspreads(thatis,thecostofborrowing)byanaverageof20%[24].
Anotherimportantaspectisthecostreductioninpolicymaking-replacingorsubstitutingtraditionaldatacollectionandevaluationmethods.
Thelargeamountofdatareadilyavailablewillenablemoretimelyanalysisofpolicyinterventions.
6.
BigDataandPolicyforthe2030AgendaforSustainableDevelopmentTherefore,thesectionbelowissuggestingactionablestepsthatcanbetakenbypolicystakeholders.
Certainly,itisnotsuggestedtodismissothermethodsofgaininginputsforpolicymaking.
Also,bigdataisnotnecessarilyforpolicymakingineverysector.
Datadrivenmethodsareespeciallybeneficialforpolicyareaswithlargevolumesofdata-suchashealth,macroeconomics,transport,migrationandtheenvironment.
TheavailabilityofbigdataprovidesauniqueopportunitytosupporttheachievementsoftheSDG'slikeneverbefore.
Asthepost-2015developmentagendahasnowbeenestablished,strengtheningdataproductionandtheuseofbetterdatainpolicymakingandmonitoringarebecomingincreasinglyrecognizedasfundamentalmeansfordevelopment.
TheMDGmonitoringexperiencehasclearlydemonstratedthateffectiveuseofdatacanhelptogalvanizedevelopmentefforts,implementsuccessfultargetedinterventions,trackperformanceandimproveaccountability.
Thus,thesustainabledevelopmentdemandsadatarevolutiontoimprovetheavailability,quality,timelinessanddisaggregationofdatatosupporttheimplementationofthenewdevelopmentagendaatalllevelsinallregions.
BigDataisanessentialpartofthedatarevolution,andchaptersixbelowidentifiespotentialareaswhereUN-ESCAPcanplayavitalroletosupportpolicymakingusingbigdataforsustainabledevelopment.
LocalizingtheSDG'sbasedonlocalprioritieswillbekeytomakethemtangibleandrelevanttargets.
6.
1.
AVisionforBigDataandthe2030AgendaAsdescribedaboveinchapterthreeonecosystems,theverynatureofbigdatarequiresnewformsofinter-institutionalrelationshipsinordertoleveragedataresources,humantalent,anddecision-makingcapacity.
Thenecessarycapabilitiesenabletheintegrationofbigdataintoongoingpolicyprocessesratherthanone-timepolicydecisions,therebyenablingitsvaluetobecontinuallyreleasedandrefined.
Spaceswillbeneededinwhichtechnical,cultural,andinstitutionalcapabilitiescancommensuratelydevelop.
Giventhevarietyandpervasivenessofthenecessarycapabilitiestoutilizebigdatatoaddressbigproblems,collaborativespacesareneededtoenhancethecapacityofindividuals,organizations,businessesandinstitutionsto31elucidatechallengesandsolutionsinaninteractivemanner,strengtheningaglobalcultureoflearning.
Someelementsofthisnewecosystemarealreadyemerging.
TheUNStatisticalCommissionestablishedaglobalworkinggroup(GWG)mandatedtoprovidestrategicvision,directionandcoordinationofaglobalprogramonBigDataforofficialstatistics[23].
Thegroupfoundthatnontraditionalsourcesofdata,especiallybigdatathatthusfarhavebeenunderutilizedinproducingofficialstatistics.
BigDatasourcesneedtobeleveragedandconsideredforadequacytoenrichthesourcesofofficialstatisticssothatthedataneedsinnewdevelopmentareascanbesatisfiedandtimely,detailedandspatiallydisaggregateddatacanbeproducedandmadeavailabletodecisionmakers.
Thisimpliesthattheinnovativeandtransformativepowerofinformationtechnologymaybeharnessed:fromthecollectionstage(through,forexample,theuseofcomputer-assistedcollectionsthroughmobiledevices),tothedisseminationstage(throughadvancedvisualizationtools,suchasdataonmaps).
6.
2.
PossibleActionTheUN'sSecretary-General'sIndependentExpertAdvisoryGrouponaDataRevolutionforSustainableDevelopment(IEAG)iscallingforactiontomobilizethedatarevolutionforsustainabledevelopment[5].
Therecommendationsoutofthisgrouprelevantforbetterpolicymakinghavebeentakenasabasisforapossibleactionstepsdescribedbelow.
UNESCAPisinauniquepositiontosupportemerginggroupsandnetworks,leveragingexistingknowledgeandresources.
Itcanfacilitatedialogueandbringtechnicalexpertiseintotheconsultations-agreeingonconcreteactionsandsharedresponsibilitiesamongallstakeholders.
Inordertogainthemaximumbenefitthatbigdataofferstothepolicymakingprocessandsustainabledevelopment,itisimportantthatvariousaspectsareaddressedinaholisticway.
Theseaspectsincludehighlevelstakeholderagreementsaswellasaccesstoinnovationsandcapacitystrengthening.
BasedontherecommendationsoftheUNIndependentAdvisoryGrouponaDataRevolutionofSustainableDevelopment[4]andUN'sGlobalPulse[41],thefollowingstepsarerecommendedasaninitialsuggestedregionalimplementationroadmap:1.
EstablishandmanageacoordinationmechanismwiththekeyUNstakeholders,(GlobalWorkingGroup(GWG)onBigDataforOfficialStatistics,GlobalPulseanditsregionaloffices)andotherinternationalpartners.
Resourcesarelimited,andaneffectivecoordinationcanenhanceknowledgesharing,fasterreplicationofinnovationsandadvanceprogressinanovelway.
2.
DevelopaconsensusonprinciplesandstandardsamongtheUNESCAPmembercountries.
Thiswouldincludeaparticipatoryandinclusiveseriesofstakeholdermeetings,bringingtogetherthepublic,privateandcivilsocietytobuildtrustand32confidenceamongdatausers.
Thiscanthenfeedintothe"GlobalConsensusonData"tobefacilitatedbytheUN.
Inaddition,UNESCAPgovernmentsshouldbebroughttogetherasasubgroupofstakeholders-focusingspecificallyontheuseandavailabilityofdataforpolicymaking.
3.
KickoffandInstitutionalizeaRegionalMulti-StakeholderMechanismtoshareinnovations.
ThePulseLabJakartacouldsupportthiseffort.
Theultimatemechanismcanbeadigitalnetwork,equivalenttoaUNESCAPCommunityofPractice(CoP)onBigDatabasedPolicyInnovations.
ThisCoPcouldthenalsoleadtheidentificationofspecificareasofinnovationfocus(e.
g.
incentives,researchetc.
)todefineLocalTangibleBenefits-BigDatashouldnotbeanendinitself.
WhiletheBigDataisaninterestingareatoexploreinitself,itisimportanttobearinmindthattheapplicationofbigdatainthepolicymakingdiscourseshouldultimatelybenefitthepeopleoftheAsiaPacificRegionbyachievingtheSustainableDevelopmentGoals.
Itisnotlikelythatprojectsusingdatawillgetthisrightthefirsttime.
Itwillbeamatteroftesting,re-testing,adjustingandlearning.
Thepointhereisnottoexperimentalldayinboutiquelabswithlittleregardtoimpact,butrathertointegrateexperimentationandadaptationattheheartofhowweimplementatscale.
toinitiatetheLocalizationofSDG'sindicators.
ThebigdatadiscourseinthiscontextcanbeusedtoengagePartnerGovernmentsinthedraftingoftargetsandtoincludeallrelevantstakeholders.
4.
MobilizeregionalresourcesforcapacitydevelopmentforthelessadvancedUNESCAPmembercountriesWhilebigdataisavailableinalmostunlimitedamounts,thecapacitytoactuallyusethedataandtofeeditintopolicymaking,islimited,especiallyinmanyoftheUNESCAPmembercountries.
Whilesomegovernmentsarequiteadvancedintheutilizationofbigdata,suchasSouthKoreaandSingapore-othercountriesbarelybenefitfromthedatarevolution.
Acapacitydevelopmentapproachbasedonpeertopeerlearning,connectedtotheabovementionedCoP'sshouldbefurtherdiscussed.
Also,availableresourcesshouldbemapped.
Inaddition,UNESCAPshouldworkwithitsnetworkstomobilizeadditionalresources.
5.
Enhancein-housebigdataanalyticscapacity.
ThegeographiccoverageofUNESCAPishugeandasupportneedamongitsmembersvariessignificantly.
Additionaltechnicalresourcesareneededtomaintainthemomentumandmakethebigdatarevolutionhappen.
UNESCAPtoestablishandmanageaRegionalSustainableDevelopmentBigDataPolicySecretariatincollaborationwiththeGlobalPulseandPulseLabJakarta.
Thissecretariatcouldleadtheabovementionedstakeholdersmechanism,provideanalyticalcapacitystrengthening,mobilizethestakeholdersandcoordinatetheproposedactions.
33TheseinitialrecommendationsaresupposedtostimulateadiscussionduringtheBigDataandthe2030AgendaforSustainableDevelopment:AchievingtheDevelopmentGoalsintheAsiaandthePacificRegionWorkshopinBangkokon14-15December2015.
34References[1]Carr-Hill,R.
(2013)'Missingmillionsandmeasuringdevelopmentprogress',WorldDevelopment46:30-44.
[2]Granoff,I.
etal.
(2014)'Targetingzero:achievingzeroextremepovertyonthepathtozeronetemissions.
'London:OverseasDevelopmentInstitute.
[3]Elizabeth,Stuart,andOthers(2015).
Thedatarevolution.
Findingthemissingmillions.
ODIDevelopmentprogress.
[4]HelenMargetts(2013);inhttp://blogs.
oii.
ox.
ac.
uk/policy/promises-threats-big-data-for-public-policymaking/[5]UnitedNations(2014),p.
6.
AworldthatCounts:MobilizingaDataRevolutionforSustainableDevelopmentbytheIndependentExpertAdvisoryGrouponaDataRevolutionforSustainableDevelopment.
NewYork.
[6]UNECEStatisticsWikis.
HowBigisBigData.
http://www1.
unece.
org/stat/platform/pages/viewpage.
actionpageId=99484307.
[7]WorldBank(2014).
BigDatainActionforDevelopment.
CentralAmerica.
[8]DineshMavaluru,etal.
(2014).
BigDataAnalyticsinInformationRetrieval:PromiseandPotential.
Proceedingsof08thIRFInternationalConference.
Bengaluru,India.
[9]K,Arun;L.
Jabasheela(2014).
BigData:Review,ClassificationandAnalysisSurvey.
InternationalJournalofInnovativeResearchinInformationSecurity(IJIRIS)ISSN:2349-7017(O).
Volume1Issue3.
[10]EmmanuelLetouzé(2015).
BigDataandDevelopment:AnOverview.
Data-PopAlliancePrimersSeries.
[11]EmmanuelLetouzé(2012).
BigDataforDevelopment:WhatMayDetermineSuccessorfailureOECDTechnologyForesight,Paris.
[12]Christopher,Surdak(2014).
DataCrush:HowtheInformationTidalWaveisDrivingNewBusinessOpportunities.
Amacom[13]Hellerstein,Joe(9November2008).
ParallelProgrammingintheAgeofBigData.
GigaomBlog.
[14]Segaran,Toby;Hammerbacher,Jeff(2009).
BeautifulData:TheStoriesBehindElegantDataSolutions.
O'ReillyMedia.
P257.
[15]Hilbert,Martin;López,Priscila(2011).
TheWorldtechnologicalcapacitytoStore,Communicate,andComputeInformation.
Science332(6025).
P60-65.
[16]IBM(2013).
WhatisBigDataBringingBigDatatotheEnterprise.
www.
ibm.
com.
[17]http://www.
kdnuggets.
com/2015/04/interview-emmanuel-letouze-democratizing-benefits-big-data.
html.
[18]GaryKing(2013).
https://gking.
harvard.
edu/files/gking/files/evbase-gs.
pdf.
InstituteforQuantitativeSocialScience,HarvardUniversity.
TalkattheGoldenSeedsInnovationSummit,NewYorkCity.
[19]UnitedNationsGlobalPulse(2014).
MiningIndonesiaTweetstoUnderstandFoodPriceCrises.
NewYork.
[20]UnitedNationsGlobalPulse(2015).
BlogDataPhilanthropy:WhereAreWeNowNewYork35[21]UnitedNationsGlobalPulse(2013).
http://www.
unglobalpulse.
org/data-philanthropy-where-are-we-now.
NewYork.
[22]JoelGurin;in:http://www.
theguardian.
com/public-leaders-network/2014/apr/15/big-data-open-data-transform-government.
[23]UnitedNations(2014),p.
6.
AworldthatCounts:MobilizingaDataRevolutionforSustainableDevelopmentbytheIndependentExpertAdvisoryGrouponaDataRevolutionforSustainableDevelopment.
NewYork.
[24]Martijn,Poel,andothers(2015).
DataforPolicy:Astudyofbigdataandotherinnovativedata-drivenapproachesforevidence-informedpolicymaking.
Brussels.
[25]JoelGurin;in:http://www.
theguardian.
com/public-leaders-network/2014/apr/15/big-data-open-data-transform-government.
[26]Thomas,J.
,andM.
Grindle.
(1994).
"PoliticalLeadershipandPolicyCharacteristicsinPopulationPolicyReform.
"PopulationandDevelopmentReview20Supp:51–70.
[27]http://www.
policyproject.
com/policycircle/content.
cfma0=4[28]LuisCrouch(2015).
ARelevantDataRevolutionforDevelopment.
RTIPressInternational.
NewYork.
[29]https://texaspolitics.
utexas.
edu/archive/html/bur/features/0303_01/policy.
htmlTheTexasPoliticsProjectattheUniversityofTexasatAustin,USA.
[30]Hayes,M.
T.
(2001)TheLimitsofPolicyChange:Incrementalism,Worldview,andtheRuleofLaw(WashingtonD.
C.
:GeorgetownUniversityPress).
[31]Isaacs,S.
andIrvin,A.
1991.
PopulationPolicy:Amanualforpolicymakersandplanners,SecondEdition.
NewYork:ThedevelopmentLawandPolicyProgram,CenterforPopulationandFamilyHealth,ColumbiaUniversity,andFuturesGroup.
[32]ChrisYiu(2013).
TheBigDataOpportunityMakinggovernmentfaster,smarterandmorepersonal.
PolicyExchange.
[33]William.
A.
Brantley(2012).
AgilePolicyMaking:HowComplexityTheory,BigDataandDataScienceResearchisChangingThePracticeofPolicyMaking.
AmericanSocietyforPublicAdministration.
[34]Steven,Vanroekel(2013).
PortfolioStat2.
0:DrivingBetterManagementandEfficiencyinFederalIT.
WhiteHouse,USA.
[35]Joel,Gurin,andLaura,Manley(2015).
OpenDataforSustainableDevelopment.
WorldBank.
[36]Kenneth,Neil,Cukier,andViktor,Mayer-Schoenberger(2013).
TheRiseofBigData.
ForeignAffairs.
[37]Jackie,Hoi-Wai,Cheng,NationalEconomistofUNDPChinaUNDP(2014).
BigDataforDevelopmentinChina.
UNDPChina.
[38]DataForPolicyastudyofbigdataandotherinnovativedata-drivenapproachesforevidence-informedpolicymaking.
http://www.
data4policy.
eu/#!
appendixb/c1kb8(2015).
TheworkshoponData-DrivenInnovationsforBetterPolicies,Brussels.
[39]RockefellerFoundationBellagioCentreconference,(2014).
Bigdataandpositivesocialchangeinthedevelopingworld:AwhitepaperforpractitionersandresearchersOxford:OxfordInternetInstitute.
[40]UnitedNations(2012).
TheFutureWeWant.
OutcomeDocumentoftheUnitedNationsConferenceonSustainableDevelopment.
RiodeJaneiro,Brazil.
36[41]GlobalPulse(2012).
BigDataforDevelopment:Challenges&Opportunities.
NewYork.
[42]ShailendraKumar(2014).
TheDatawithinBigDataandthemytharoundUnstructuredData.
[43]UNECEStatisticsWikis(2013).
http://www1.
unece.
org/stat/platform/display/bigdata/Classification+of+Types+of+Big+Data.
[44]Judith.
Hurwitz,etal(2013).
BigDataForDummies.
JohnWiley&Sons,Inc.
37Annex1:BigDataTypesVarietyisoneoftheprinciplesofBigDataasdescribedpreviously.
TheBigDatacanbedividedintothreetypes[42,43,44]:StructuredData,Semi-StructuredData,andUnstructuredData.
Definitionsandexamplesofeachcanbedescribedasfollows:StructuredDataStructureddatagenerallyreferstodatathathasadefinedlengthandformat.
Mostorganizationsarestoringlargeamountsofstructureddatainvariousdivisions,innormalised/deformalisedformatsinadatabase:Datawarehouses,relationaldatabasemanagementsystem(RDMSs),andvariousotherenvironments.
Thedatacanbequeriedusingalanguagelikestructuredquerylanguage(SQL)inwhichthedatasetscanbeupdatedwithnewdata,anddeleted,readoranyotheractivity.
Theevolutionoftechnologyprovidesnewersourcesofstructureddatabeingproduced-ofteninrealtimeandinlargevolumes.
Thesourcesofdataaredividedintothreecategories:(a)Computer-orMachine-GeneratedStructuredDataMachine-generateddatagenerallyreferstodatathatiscreatedbyamachinewithouthumanintervention.
Theycanincludethefollowing:Sensordata:ExamplesincluderadiofrequencyID(RFID)tags,smartmeters,medicaldevices,andGlobalPositioningSystem(GPS)data.
AnotherexampleofsensordataissmartphonesthatcontainsensorslikeGPSthatcanbeusedtounderstandcustomerbehaviorinnewways.
Forexample,RFIDisrapidlybecomingapopulartechnology.
Itusestinycomputerchipstotrackitemsatadistance.
Anexampleofthisistrackingcontainersofproducefromonelocationtoanother.
Wheninformationistransmittedfromthereceiver,itcangointoaserverandthenbeanalyzed.
Companies,forexample,areinterestedinthisforsupplychainmanagementandinventorycontrol.
Weblogdata:Whenservers,applications,networks,etcoperate,theycaptureallkindsofdataabouttheiractivity.
Thiscanamounttohugevolumesofdatathatcanbeuseful,forexample,todealwithservice-levelagreementsortopredictsecuritybreaches.
Point-of-saledata:Whenthecashierswipesthebarcodeofanyproductthatyouarepurchasing,allthatdataassociatedwiththeproductisgenerated.
Just38thinkofalltheproductsacrossallthepeoplewhopurchasethem,andyoucanunderstandhowbigthisdatasetcanbe.
Financialdata:Lotsoffinancialsystemsarenowprogrammatic;theyareoperatedbasedonpredefinedrulesthatautomateprocesses.
Stocktradingdataisagoodexampleofthis.
Itcontainsstructureddatasuchasthecompanysymbolanddollarvalue.
Someofthisdataismachinegenerated,andsomeishumangenerated.
(b)Human-GeneratedData:Thisisdatathathumans,ininteractionwithcomputers,supply.
Inputdata:Thisisanypieceofdatathatahumanmightinputintoacomputer,suchasname,age,income,non-free-formsurveyresponses,andsoon.
Thisdatacanbeusefultounderstandbasiccustomerbehavior.
Click-streamdata:Dataisgeneratedeverytimeyouclickalinkonawebsite.
Thisdatacanbeanalyzedtodeterminecustomerbehaviorandbuyingpatterns.
Gaming-relateddata:Everymoveyoumakeinagamecanberecorded.
Thiscanbeusefulinunderstandinghowendusersmovethroughagamingportfolio.
Thewaydataisstructuredisavitalelement.
Ifthestructuresaren'tcoherentandunderstandable,dataisliabletobemisused(misunderstood)andwillfailtofacilitate"bringingtogether"datafromdisparatesourcestoproducenewknowledge/evidence.
Thisisametadataschemarelatedissue–orbroughtdowntoasimpleexamplewhatheadings/termsarebeingusedforcolumnsofdatainaspreadsheetandhowcanthepersonusingthespreadsheetunderstandthecontext.
Semi-StructuredDataSemi-structureddataisakindofdatathatfallsbetweenstructuredandunstructureddata.
Thistypeofdatabecameatalkingpoint.
MostlydatacomingfromFacebook,Twitter,Blogs,publicallyavailablewebsites,etc.
makesthebasisofsemi-structureddata.
Thesedatasourcesusuallyhavedefinedstructuresandmostlycontaintextinformation.
Thefreeflowtextgeneratedthroughthesocialmediaistheonlyunstructuredcomponentwhilsttheremainingdataisstructured.
Mostofthetimes,thesocialdataismistakenwithunstructureddata.
ThesocialdataisNOTunstructureddata,itissemi-structuredandinfact,someofthesocialdatacontainsindustrystandardstructures.
Socialmediadata:ThisdataisgeneratedfromthesocialmediaplatformssuchasYouTube,Facebook,Twitter,LinkedIn,andFlickr.
39UnstructuredDataDataUnstructureddatadoesnothaveanydefined,consistentfieldsanditmayevendonothaveanynumbersandtext.
Unstructureddatacanbedividedalsointoeithermachinegeneratedorhumangeneratedanddescribedasflows:(a)Machine-GeneratedUnstructuredDataExamplesSatelliteimages:Thisincludesweatherdataorthedatathatthegovernmentcapturesinitssatellitesurveillanceimagery.
JustthinkaboutGoogleEarth,andyougetthepicture(punintended).
Scientificdata:Thisincludesseismicimagery,atmosphericdata,andhighenergyphysics.
Photographsandvideo:Thisincludessecurity,surveillance,andtrafficvideo.
Radarorsonardata:Thisincludesvehicular,meteorological,andoceanographicseismicprofiles.
(b)Human-generatedUnstructuredDataExamplesMobileandVoicedata:Thisincludesdatasuchastextmessagesandlocationinformation.
Humanvoicecontainsalotofinformationanditneedsaccessandmined.
Thespectrogramofthehumanvoicerevealsitsrichharmoniccontentincludingpitch,tone,emotion,bass,etc.
Webbehaviorandcontent:Thiscomesfromanysitedeliveringunstructuredcontent,likeYouTube,Flickr,orInstagram.
Thescopeofwebbehaviorishuge.
TherearenearlyfivebillionindexedwebpagesontheInternetandforeachpagetherearetrafficstatisticsrangingfromthenumberanddurationofvisitstofarricherinformationonuserbehavioronalargeproportionofwebsites.
BigDataalsoencompassesthecontentofthosewebpagesandthechangesthatoccuronthem.
Alsoincludedinthiscategoryisthevastamountofsearchenginedataconstantlybeinggenerated.
ImageandVideoData:Totalnumberofpicturestakeninlast5yearsismorethandoublethepicturestakenin1900-2000.
Thisgivesusanopportunitytousepatternswithinthepicturesandminetheinformationavailabletous.
Varioustechniqueslikepixilation,patternmatching,imageprocessing,featureextracting,etc.
allowsuscovertthepicturesintodataandfurthermineitusingclassificationalgorithms.
Examplesofimagedatausecases:Oneofthemost40commonusecaseisthethumbprintrecognitionwhichisnowavailableinourphonesandonelargebankisusingtheimageminingtechniqueandpredictinglikelihoodofacustomertobefraudwhileincaseofthevideoOneverylargesecurityagencyusesthevideodatatoidentifytroublemakingcandidatesinthepremisebyusingpredictiveanalysisbasedonthesequenceofactionsperformedbytheindividual.
MachineData:Asthesizeofcomputerchipsisreducing,thereispotentialofhavingacomputerchipinalmostallthemachines,e.
g.
cars,themobilephones,ships,etc.
Thedataresidinginthesemachinesisunstructuredandisnotofastandardformattobeavailableformining.
Thisunstructureddataisbeingextractedbylargeorganisationsandthenusedtounderstandthehiddenpatternstodriveefficiency.
Exampleofmachinedatausecase:AlargeTelcointheUSisusingmobileappdatatoadvertiseandpromoteretailoffersbyunderstandcustomerbehaviourandAlargecarcompanyiscollectingdatafromthecarstounderstandthereasonbehindenginefailuretooptimisetheperformanceandreduceenginefailurepossibilities.
41Annex2:BigDataCaseStudies1.
CaseStudyIUSINGFLOWMINDERTOFOLLOWPOPULATIONDISPLACEMENTAFTERTHENEPALEARTHQUAKEINAPRIL2015Flowminder.
orgdevelopedatooltoprovidekeyinformationonlargescaledisplacementtakingplaceaftertheNepaldisaster.
Throughtheuseofanonymousmobileoperatordatatheywereabletomeasureandvisualizepopulationmovementsandthisresultedinmoreequitablesupporttopeoplestruckbytheearthquakeregardlessoftheirlocation.
PROBLEM:AsiaPacificisthemostdisasterproneregionoftheworld.
Annually,millionsofpeopleremainatrisktoearthquakes,tsunamis,tropicalcyclones,typhoons,floodsandstormsurges1.
Thepooraremoreimpactedbynaturaldisastersbecausetheyaremorevulnerableandusuallytheirlivelihoodsdependonclimateandlandbasedsubsistence.
Theyarealsolesslikelytohavesocialprotections,insurance,orcapacitytorecoverafteradisaster.
Thus,disasterandriskreductionpoliciesandmeasuresshouldbeincorporatedinpovertyreduction,developmentandenvironmentalstrategiestocreatemoredisasterresilientsocietiesandcommunities,facingdecreasedlevelofriskandvulnerability.
Followingmajordisastersthereisapatternofpopulationmovement,andtwentytothirtymillionpeoplearedisplacedduetonaturaldisasterseveryyear.
Inmostcases,traditionaltoolsusedindisasterresponseandpreparedness-includingeyewitnessaccounts,manualcountingofpeople,registrationincampsorsatelliteoraerialimagesofsheltersorchangesinvegetation2-arenotabletodocumentinatimelyandaccuratemanner.
Predictingandmonitoringpopulationdisplacementcanreducethepopulation'svulnerabilityandhelpprovidetargetedreliefassistanceandpreventdiseases.
USINGBIGDATATOUNDERSTANDAFFECTEDPOPULATIONMOVEMENTSDURINGADISASTER:Asopposedtotraditionaldisasterresponseandpreparednesstools,utilizingBigDataandnewtechnologiescanofferanexcellentalternativetomapaffectedpeopleandtheirmovements.
Flowminder3workswithlargemobileoperator'sdatabases.
TheunderlyingtechnologythatFlowminderisusingreferstogeographicpositionsofSIMcardswhicharedeterminedbythelocationofthemobilephonetowerthroughwhicheachSIMcardconnectswhencalling.
Throughanalysisofthesedatasets,Flowminder1ESCAPTrustFundforTsunami,DisasterandClimatePreparednessBrochure2BengtssonL,LuX,ThorsonA,GarfieldR,vonSchreebJ(2011)ImprovedResponsetoDisastersandOutbreaksbyTrackingPopulationMovementswithMobilePhoneNetworkData:APost-EarthquakeGeospatialStudyinHaiti.
PLoSMed8(8):e1001083.
doi:10.
1371/journal.
pmed.
10010833FlowminderFoundationisanon-profitorganizationwithamissiontoprovideglobalpublicgoodsthroughthecollection,analysisandintegrationofanonymousmobileoperator,satelliteandhouseholdsurveydata42canmapthedistributionsandcharacteristicsofvulnerablepopulationsinlowandmiddleincomecountries.
Followingthedevastating7.
8magnitudeearthquakeon25April2015inNepalintheGhorkadistrictwhichkilledmorethan9,000peopleandcausedinjuriestomorethan23,000,FlowmindersupportedtheNepaliGovernment,UnitedNationsentitiesandotherreliefagencieswithdisplacementanalyses.
FlowminderenteredintoapartnershipwithNcell,thelargestmobileoperatorinNepaltohaveaccesstotheanonymizeddataof12millionphones.
Asshowninthe,thepre-earthquakepopulationwas2.
8millionwithabnormaloutflowsfromtheKathmanduValleytootherdistrictsof390,000people4,5.
KEYPLAYERS:NepaliGovernment,Flowminder,UNreliefagencies,NCellOUTCOMES:ThedatathatFlowmindergatheredandanalyzedwiththecontributionofNcellweresharedwithdifferentUNandnon-UNreliefactorssuchastheUNOfficeforCoordinationofHumanitarianAffairs(OCHA),UNWorldFoodProgram,andtheInternationalOfficeforMigration.
Theinformationwasusedtoplanaiddistributionandestimatethenumberofpeopleaffected.
Organizationscanusethereal-timedatatounderstanddisplacementmechanismsanddeveloptargetedsystemsforprovisionofreliefresponse6.
InthecaseofNepalFlowminder,throughtheiranalysis,foundthatafteradjustingfornormalmovementpatterns,whichwouldhavetakenplaceintheabsenceoftheearthquake,anestimatedadditional500,000peoplehadlefttheKathmanduValleytwoweeksaftertheearthquake.
ThemajorityofthesewenttothesurroundingdistrictsandtheTeraiareasintheSouthandSoutheastofNepal(FlowminderNepalCasestudy).
ThoughanalysisoftheNepalresearchresultsisongoing,apreviousstudyconductedbytheFlowminderteamaftertheHaitiearthquakein2010showedthattherewasacorrelationofdisplacedpeople'sdestinationtowheretheyhadsignificantsocialbonds7.
BigDataofferunprecedentedinsightintohumanbehaviorthatisunparalleledtothepreviousmethodsenlistingsurveysandstaticmethodstocollectself-reportedindicationsofaction.
CONCLUSIONS:NaturalDisastersareamajorthreattoSustainableDevelopment.
Somecountries,especiallythosecountrieswithspecialneeds,donotyethavethemechanismsinplacetoprovideeffectivedisasterresponseandpreparedness.
Asacountryhighlypronetonaturaldisasters,apriorityforNepalisdevelopmentofpoliciesandpracticesthatemphasizedisasterresilienceandpreparednesstominimizetheimpactonpovertyeradicationandsustainabledevelopmentefforts.
BigDataoffersanopportunitytoenhanceearlywarningsystems,strengthenresilienceandensure4NepalEarthquake2015,FlowminderCaseStudy(2015)5NcellPicture:Accessibleathttp://i.
imgur.
com/xnGbX92.
jpg16NepalEarthquake2015,FlowminderCaseStudy(2015)7BengtssonL,LuX,ThorsonA,GarfieldR,vonSchreebJ(2011)ImprovedResponsetoDisastersandOutbreaksbyTrackingPopulationMovementswithMobilePhoneNetworkData:APost-EarthquakeGeospatialStudyinHaiti.
PLoSMed8(8):e1001083.
doi:10.
1371/journal.
pmed.
100108343efficientandeffectiveactionafteradisasterhasoccurredtolimitdamage.
Morespecifically,theutilizationoftechnologycanmassivelyincreasetheefficiencyofprovisionofaidandbetterstructuresforreliefresponsefordisplacedpopulations.
TherealtimepredictivemechanismthatFlowminderisusingthroughtheanalysisoflargedatasetscangiveaninsightinpopulationdisplacementimmediatelyfollowinganaturaldisaster.
Thetypeofinsightsthatwecandrawfromlargedatasetsshedlighttohumanbehaviorregarding1)mobility,2)socialinteractionand3)economicactivity8.
Thesedatacanhelppolicymakersidentifytheappropriatepolicystrategiesassociatedwithdisasterresponse.
Knowingwherethedisplacedpopulationsareamassedcanleadagenciestobettertargetpoverty-alleviationpoliciessuchasfoodandnutrition,unemploymentassistanceandmicrofinance9.
Itisclearthatthereisaroleformulti-stakeholderpartnershipstodeliveronthepotentialoftheuseofbigdataindisasterpreparednessandresponse.
InthecaseofNepal,theprivatesector(NCell)gotinvolvedandafterthedevastatingearthquaketheyutilizeddata-sharingforsocialgood.
DatasharingorDataPhilanthropyisessentialtoensurefreeaccesstolargedatasetswhichcanbeusedtoimprovepublicpolicies.
PreparedbyErifyliNomikou,Consultant,EDD/ESCAPReferences:[1]ESCAPTrustFundforTsunami,DisasterandClimatePreparednessBrochure[2,7]BengtssonL,LuX,ThorsonA,GarfieldR,vonSchreebJ(2011)ImprovedResponsetoDisastersandOutbreaksbyTrackingPopulationMovementswithMobilePhoneNetworkData:APost-EarthquakeGeospatialStudyinHaiti.
PLoSMed8(8):e1001083.
doi:10.
1371/journal.
pmed.
1001083[3]FlowminderFoundation.
AboutuspageAccessible:http://www.
flowminder.
org/about[4,6]NepalEarthquake2015,FlowminderCaseStudy(2015)[5]PictureCreditsNcell:Accessibleat:http://i.
imgur.
com/xnGbX92.
jpg1[5]Shakya.
A,29May-5June2015#760"Wherearewe:NcellpartnerswithFlowmindertotrackmovementofNepalispost-earthquake"Accessibleat:http://nepalitimes.
com/article/nation/Ncell-Flowminder-track-movement-of-nepalis-post-earthquake,2278[8]UnitedNationsGlobalPulse(October2013)MobilePhoneNetworkDataforDevelopment.
[9]OECDPolicyDevelopmentCenter,NaturalDisasterandVulnerability,Policybriefing29Accessible:http://www.
oecd.
org/dev/37860801.
pdf8UnitedNationsGlobalPulse(October2013)MobilePhoneNetworkDataforDevelopment.
9OECDPolicyDevelopmentCenter,NaturalDisasterandVulnerability,Policybriefing29442.
CaseStudyIIUSINGBIGDATATOSUPPORTE-WASTEMANAGEMENTINCHINABaiduRecycle,awebbasedapplicationlaunchedbyUNDPandBaidu,helpstoproperlydisposee-wasteinChina.
Chinaisthesecondbiggeste-wasteproducerandbiggeste-wasteimporter.
Theapplicationhassuccessfullybeenusedtocollectandrecycle11,429electronicitemssinceitsinceptioninAugust2014.
InNovember2015,theBaiduRecycleGreenServiceAlliancewasestablishedbyBaiduandUNDPtofurtherhelptheAppscaleupandpromoteandinternet–basednationwidee-wastemanagementecosystem.
PROBLEM:Asia-Pacificisamongtheworld'stopregionsgeneratingandimportinghighlevelsofelectricalandelectronicequipmentwasteore-wasteforshort.
E-wastecoversitemsofalltypesofelectricalandelectronicequipment(EEE)anditspartsthathavebeendiscardedbytheowneraswastewithouttheintentionofre-use10.
Chinaisthesecondbiggeste-wasteproducerandbiggeste-wasteimporter11.
Chinesenationale-wastegrewfrom2009to2013atanannualaverageof21.
6%,outofthe3.
6milliontonsofe-wastebeinggenerateddomestically,onlyabout40%wereprocessedbyformalchannels12.
Theinformalsectorplaysanimportantroleinthecollectionanddisposalofe-wasteinChinaandotheremergingcountries.
Ensuringsustainableconsumptionandproductionpatterns(SDG12)isimportantforpropere-wastemanagementandresourceefficiency.
Acommitmentfrommemberstatestointernationalregulationandtechnicalstandardswouldenhanceenvironmentalsustainability,ensurethatpreciousandscarceresourcesarenotlostandleadtohealthierenvironmentswhichpromotethehumanwell-being(SDG3).
USINGBIGDATATOENSUREPROPERE-WASTEMANAGMENT:Technologycanfuelinnovativewaystomanagee-wasteandcreateresponsiblerecyclingbehavior.
BaiduInc.
,aleadingChinesecompanyinwebservices,andUNDPenteredintoastrategicpartnershiptoco-createaBigDatajointlaboratory.
Beyonditsprimaryfocusonenvironmentalissues,thelaboratorywillexploretheuseofBigDatatechnologiestosolveotherglobalproblemssuchashealth,educationanddisasters.
ThefirstproductlaunchedbytheUNDP-BaiduJointBigDataLabwasawebbasedapplicationaimingtoimprovemonitoringofe-wastedisposalandrecyclingbehavior,andraiseawarenessaboutenvironmentallyappropriateapproachestoe-wastedisposalthroughproceduresthatdonotfallintheinformalmarket.
TheusersdonotneedtodownloadanappbutratheruseapictureoftheirelectronicdeviceonBaidu'sRecyclesearchapp.
Theresultofthatresearchyieldsname,typeandestimatedvalueoftheelectronicitem.
Thenuserscanarrangeforadoor-todoore-wastepickup.
Thesuccessofthefirstversionofthiswebbasedappledtotheexpansionoftheappsservice13.
InitiallythecoverageoftheresearchdatabasesincludedonlyTVs,washingmachines,refrigeratorsanddigitalproductswhichexpandedtoincludecell-phonesandlaptops.
10SolvingtheE-WasteProblem(Step)InitiativeWhitePaper.
"OneGlobalDefinitionofE-waste".
(2014)11Cheng,J.
UNDPChinaworkingPaper.
"BigDataforDevelopmentinChina"(2014)12UNDP.
"HarnessingthePowerofBigData"(2014)13UNDP.
"China:TurningE-TrashintoCash".
(2015)45KEYPLAYERS:GovernmentofthePeople'sRepublicofChina,UNDP,BaiduInc.
INSIGHTS&OUTCOMES:Baiduhasbeenvastlysuccessfulinhelpingtodevelopintelligentsolutionsfore-wasterecycling.
Usingphotographstomatchelectronicequipmentacrossdifferenttypeswithdatasetsisaninnovativewaytoallowcustomerstosharetheirdisposalneedswhilecreatinganefficientmanagementofe-waste.
InAugust2015,accordingtoUNDP,11,429electronicitemshavebeensuccessfullyrecycledandtreated,370,000pageviewsofBaiduRecycleApphadbeenreachedandthetotaldailysearchesfortheappnumbered50,00014.
ThesedatashowthetremendouspotentialfortheRecycleapptoscaleupandreachothercitiesintheworld'smostpopulatedcountry.
TheuseoftheBaiduRecycleappactivelyrenderscitizenstodevelopgreenerrecycleconsciousandcontributetothecutdownoftheinformalrecyclestations.
Inacontinuousefforttosupportthisinitiative,inNovember2015,BaiduandUNDPlaunchedtheBaiduRecycleGreenServiceAlliancetofurtherenhancetheuseofBaiduRecycleappandattractmorestakeholders.
TheAllianceaspiresinthecollaborationwithelectronicmanufacturersinordertobuildaninternet-basednationwidee-wastemanagementecosystem15.
CONCLUSION:Unsafee-wastemanagementisposingathreattoSustainableDevelopment.
Policymakersneedtoassesstheavailableopportunitiesinordertomitigateenvironmentalthreatsderivingfromimpropere-wastedisposal.
Goodpolicieswillincludeproperrecyclinginfrastructure,shiftinge-wastecollectionfromtheinformalsectortotheformal,thecreationofgreenjobsandashiftinpeople'sbehaviortowardagreenapproachtoe-wastedisposal.
EnsuringthatthesepoliciesareinplaceandmoreBigDatainitiativesintheformofPublic-PrivatePartnershipsareformedcansupporttheregiontoachieveasustainabledevelopmentfuturewhichisinclusiveforallanddoesnotlessentheenvironmentalandhealthstandards.
PreparedbyErifyliNomikou,Consultant,EDD/ESCAPReferences:[10]SolvingtheE-WasteProblem(Step)InitiativeWhitePaper.
"OneGlobalDefinitionofE-waste".
(2014).
Accessibleat:http://www.
step-initiative.
org/files/step/_documents/StEP_WP_One%20Global%20Definition%20of%20E-waste_20140603_amended.
pdf[11]Cheng,J.
UNDPChinaworkingPaper.
"BigDataforDevelopmentinChina"(2014)[12]UNDP.
"HarnessingthePowerofBigData"(2014).
Accessibleat:http://www.
cn.
undp.
org/content/china/en/home/presscenter/pressreleases/2014/08/harnessing-the-power-of-big-data.
html14UNDP.
"China:TurningE-TrashintoCash".
(2015)15UNDP.
"UNDPandBaiduLaunchedGreenAlliancetoStepupE-wasteRecyclingService"(2015)46[13,14]UNDP.
"China:TurningE-TrashintoCash".
(2015).
Accessibleat:http://www.
asia-pacific.
undp.
org/content/rbap/en/home/ourwork/development-impact/innovation/projects/china-ewaste.
html[15]UNDP.
"UNDPandBaiduLaunchedGreenAlliancetoStepupE-wasteRecyclingService"(2015).
Accessibleat:http://www.
cn.
undp.
org/content/china/en/home/presscenter/pressreleases/2015/11/04/undp-and-baidu-launched-green-alliance-to-step-up-e-waste-recycling-service0.
html3.
CaseStudyIIIUSINGMOBILEMETA-DATAFORURBANANDTRANSPORTATIONPLANNINGINPURPOSESINSRILANKASriLanka'smajorcityColomboisfacingatremendouspopulationcongestionchallenge.
LIRNEasiahasgainedaccesstohistoricalandanonymizedmobiledatatobetterunderstandpopulationmovementsinColomboCityandmakeinformedandtimelyurbanplanningandurbantransportationpolicyrecommendations.
PROBLEM:"In2010,theAsia-Pacificregion'surbanpopulationwas754millionpeople,anditisexpectedthattheurbanizationrateintheregionwillreach50percentin202616.
Makingcitiesandhumansettlementsinclusive,safe,resilientandsustainable(SDG11)isacross-cuttingissueacrosstheintegrated2030Agenda.
Thirteenoutofthetotaltwenty-twoMega-citiesarelocatedintheregion.
Populationdensityandroadcongestionareamongthedifficultiesthaturbanpopulationsarefacingasmega-citiesareexpanding.
Additionally,theimpactsofpovertyincitiesareexacerbatedbyinadequateaccommodation,slumdwellings,andunsanitaryandunsafelivingconditions.
ExistinginfrastructureisoftenunabletoaccommodatetheimpactsofthegrowthrateofAsiancities,andgrowthpatternsareleadingtounstainableconsumptionandproductionpatterns17.
Afocusonurbanplanning;urbantransportationandurbaninfrastructureisthespringboardofsustainableurbanization.
InColombo,SriLankapopulationcongestionisamajorchallengeforpublicpolicy.
47%ofthecity'sdaytimepopulationcomesfromoutsidethe16ESCAP.
"UrbanizationTrendsinAsiaPacific"(2013)Accessibleat:http://www.
unescapsdd.
org/files/documents/SPPS-Factsheet-urbanization-v5.
pdf17ESCAP.
"UrbanizationTrendsinAsiaPacific"(2013)Accessibleat:http://www.
unescapsdd.
org/files/documents/SPPS-Factsheet-urbanization-v5.
pdf47city18.
Thepopulationdensitylevel,asobservedintheheatmap,reachesitspeakduringweekdayswhenpeopleremainintheinnercityforworkorentertainment.
TheissuesthatarisefromthepopulationcongestionposethreatstothelivabilityofthecityandbyextensiontomanycitiesintheAsia-Pacificthatfacesimilarpopulationcongestionissues.
USINGBIGDATAFORURBANANDTRANSPORTATIONPLANNING:Theuseofemergingnewtechnologiesandmorespecifically,BigDatatechnologiesarecreatinganewsmartprofileforcitiesandanewtypeofcitizenshipwhichpromotessocialactivismandcitizensengagementformoreparticipatorygovernanceintheurbansystem.
DigitalurbanismortheInternetofthingsischangingtheurbanlandscapethroughtheutilizationofinformationandcommunicationtechnologies(ICTs)totackleurbanchallenges.
InSriLanka,LIRNEasia19,apro-poor,pro-marketthinktank,partneredwithmultipletelecomoperatorstogainaccesstohistoricalandanonymizedtelecomnetworkbig.
ThoseoperatorsofferedaccesstoCallDetailrecordsincludingCalls,SMSandInternetandAirtimeRechargeRecords.
ThroughtheuseofSIM-movementsdata,newinsightscanbedrawnregardinglocationandtimelineofthepopulationcongestion,origin/homelocationanddestination/worklocationandfrequencyandquantityofmobileinteractionofuserswithintheadministrativeboundariesofthecity.
Bigdataoffersacheaperandmoreeffectivealternativetotraditionallycostlycensusandhouseholdsurveystogatherinformationwhichwillbevaluableforurbanandtransportationplanning.
AtthesametimetheopportunitytoleverageBigDataishugeduetothetremendoushighcoverageofthepopulationbymobilephonesindevelopingeconomies,gaininginsightonmobilityfrequencyandgeography.
INSIGHTS&OUTCOMES:Duetotheincreaseduseofmobilephonesandthewidecoveragethattheoperatorsareofferingitiseasytoresorttomobiledatainformationforunderstandinggeographiclocations,mobilitypatternsandthefrequencyofmovementsofpopulations.
Thisinformationisextremelyusefulforpolicymakers,especiallywhenitistimely,efficientandnotcostly.
OneoftheinsightsintheColomboCitywasthatmunicipalboundariesarenolongervalid.
Intermsoftransportationpolicy,thefocusshouldturntothecreationofhighvolume18LIRNEasia,MobilenetworkbigdataforurbanandtransportationplanninginColombo,SriLanka"DataforPolicyconference,Presenters:Samarajiva,R.
Lokanathan,(2015).
Accessibleat:http://lirneasia.
net/wp-content/uploads/2013/09/Samarajiva_Cambridge_June15.
pdf19LIRNEasiamissionistocatalyzepolicychangethroughresearchtoimprovepeople'slivesintheemergingAsiaPacificbyfacilitatingtheiruseofhardandsoftinfrastructuresthroughtheuseofknowledge,informationandtechnology(LIRNEasia).
48transportationcorridorsformasstransit20.
Furthermore,theColomboDistrictwasmappedinthreespatialclustersFromthisalmostrealtimemonitoringofurbanlanduse,itconcludesthatthecentralbusinessdistrictinColombohasexpanded21.
CONCLUSION:BigDatacanplayakeyroleinachievingSustainableDevelopmentthroughthevaluableinsightsthatlargegroupsofdatacangenerateespeciallyforimprovingurbanandtransportationplanningincities.
TheresearchfindingsandrecommendationofLIRNEasia,canprovideinsightinunderstandingchangesintheurbanpopulationdensityandmobility.
Thesefindingscanhelpurbanplannersandpolicymakerstocreatemoresustainablecitiesandbenefitfromthecostsavingsassociatedwiththenewtechnologiesinsteadoftraditionallesseffectivemechanismstogatherthesedata.
TheprivatesectorofferedaccesstohistoricalandanonymizedmobiledatatoLIRNEasia.
ThiswasanopportunitytoleverageBigDatausingprivatesector'sDataPhilanthropyforpublicpolicyinsights.
Itisalsoanopportunityformobileandothercompaniestodrawinsightconcerningthepopulationtheyareservicingforcommercialandprofitmakinguses.
PreparedbyErifyliNomikou,Consultant,EDD/ESCAPReferences:[16,17]ESCAP.
"UrbanizationTrendsinAsiaPacific"(2013)Accessibleat:http://www.
unescapsdd.
org/files/documents/SPPS-Factsheet-urbanization-v5.
pdf[19]LIRNEasia.
Aboutus.
Accessibleat:http://lirneasia.
net/about/[18,20,21]LIRNEasia,MobilenetworkbigdataforurbanandtransportationplanninginColombo,SriLanka"DataforPolicyconference,Presenters:Samarajiva,R.
Lokanathan,(2015).
Accessibleat:http://lirneasia.
net/wp-content/uploads/2013/09/Samarajiva_Cambridge_June15.
pdf[21]Samarajiva,R.
"Usingmobile-networkbigdataforurbanandtransportationplanninginColombo"LIRNEasia(2015).
Accessibleat:http://www.
iesl.
lk/Resources/Documents/My%20Docs/Event%20PDF/PL%20L%2016012015.
pdf[Figure1,Figure2}Samarajiva,R.
"Usingmobile-networkbigdataforurbanandtransportationplanninginColombo"LIRNEasia(2015).
Accessibleat:http://www.
iesl.
lk/Resources/Documents/My%20Docs/Event%20PDF/PL%20L%2016012015.
pdf20LIRNEasia,MobilenetworkbigdataforurbanandtransportationplanninginColombo,SriLanka"DataforPolicyconference,Presenters:Samarajiva,R.
Lokanathan,(2015).
Accessibleat:http://lirneasia.
net/wp-content/uploads/2013/09/Samarajiva_Cambridge_June15.
pdf21LIRNEasia,MobilenetworkbigdataforurbanandtransportationplanninginColombo,SriLanka"DataforPolicyconference,Presenters:Samarajiva,R.
Lokanathan,(2015).
Accessibleat:http://lirneasia.
net/wp-content/uploads/2013/09/Samarajiva_Cambridge_June15.
pdf494.
CaseStudyIVUSINGSOCIALMEDIATOTRACKWORKPLACEDISCRIMINATIONAGAINSTWOMENININDONESIAGender-baseddiscriminationisprevalentintheAsia-Pacificregion.
Womenarepresentedwithlessemploymentopportunities,wagegapsandaremostfrequentlyvictimsofsexualharassmentatwork.
TheILOincollaborationwithPulseLabJakartausedsocialmediatoexplorewhetheronlinedatacanactasasourcefordrawingrealtimeinformationfordiscriminationagainstwomenintheworkplace.
PROBLEM:Across-cuttingissuewhichneedsurgentactionisachievinggenderequalityandempoweringallwomenandgirls(SDG5).
Overthelasttwodecades,employmentrateshaveincreasedforwomenintheregion.
Sohasthelevelofdiscrimination,notonlybasedongenderandethnicorigin,butalsoduetosexualharassment.
WomeninIndonesia,experiencelimitedaccesstoemploymentopportunitiesandtraining,andunequaltermsofemployment–bothintermsofwages,withawagegapof35percent22,aswellasintermsofprofessionalresponsibilities.
Overthepastdecade,womenparticipationratesinthelaborforcehavebeenbetween50-53percentwhileformenitisbetween80-83percent.
Historically,genderbasedworkplacediscriminationsareverydifficulttomonitorasincidentsusuallyremainunreported23.
USINGBIGDATATOUNDERSTANDDISCRIMINATIONINTHEWORKPLACE:BigDataprovidesaninnovativewaytogainusefulinsightsonpopulationbehaviorinrealtime.
InIndonesia,socialmediadataminingandmorespecifically,leveragingtweets,canbeagoodalternativetocostlytraditionalwaysofcollectingdatathroughlengthysurveystogainnewsourcesofinformationforworkplacediscrimination.
InpartnershipwiththegovernmentofIndonesiaandtheILO,theUNGlobalPulseLabinJakartatestedwhethersocialmediamonitoringcanprovidesignalsforreal-timeworkplacediscriminationagainstwomen.
TheyfilteredtweetsandextractedonlineconversationsintheBahasaIndonesialanguagefrom2010to2013.
Tweetsfallinginoneofthe8topics24werefilteredandthenanalyzedforvolumeandcontentusingasocialdataanalyticsplatformcalledCrimsonHexagontodetectwhethertheyprovidedsufficientvolumetoanalyzefurtheraspotentialsignalsofperceptions,opinionsandincidentsofdiscrimination25.
KEYPLAYERS:TheGovernmentofIndonesia,UnitedNationsGlobalPulseLabJakarta,InternationalLaborOrganization,Twitter.
22UNGlobalPulse,'FeasibilityStudy:IdentifyingtrendsinDiscriminationagainstWomenintheWorkplaceinSocialMedia",GlobalPulseProjectSeriesno11,2014'.
23UNGlobalPulse,'FeasibilityStudy:IdentifyingtrendsinDiscriminationagainstWomenintheWorkplaceinSocialMedia",GlobalPulseProjectSeriesno11,2014'.
24Thecategorieswerethefollowing:1)Permissiontowork,2)Appropriatenessofwork,3)theburdensofworkingwomen4)Discriminationinjobrequirements5)Lackofskillsoreducation6)costtoaccessemployment7)Home-basedworkersand8)sexualharassmentintheworkplace.
UNGlobalPulse,'FeasibilityStudy:IdentifyingtrendsinDiscriminationagainstWomenintheWorkplaceinSocialMedia",GlobalPulseProjectSeriesno11,2014'.
25UNGlobalPulse,'FeasibilityStudy:IdentifyingtrendsinDiscriminationagainstWomenintheWorkplaceinSocialMedia",GlobalPulseProjectSeriesno11,2014'.
50INSIGHTS&OUTCOMES:From2010to2013,socialmediainputsandonlineconversationswereanalyzeddemonstratingthatonlyfourtopicshadsufficienttweetstolookinto.
Thosewere:Permissiontowork(3,000tweets);appropriatenessofwork(5,000tweets);burdensofworkingwomen(21,000tweets)anddiscriminationinjobrequirements(78,000tweets)26.
Giventhatthevolumeofrelevantonlineconversationsisincreasingitwasconcludedthatfurtherresearchisneededandthatexistingmonitoringmechanismscouldbesupplementedbydigitaltoolstocreateadecentworkenvironment.
Theprivatesectorisanimportantplayerinofferinglargedatasetsfortheanalysis.
Morespecifically,TwitterisextremelypopularinIndonesia,andinparticularinJakarta,andpeopleareverylikelytoshareexperiencesusingtheirtwitterhandlesorothersocialmediaoutlets.
Theunderstandingofthepowerofdatasharing,toolsandexpertisefromprivatesectorisinstrumentaltothecompletionofdataprojects.
Harnessingdigitaldataforsocialgoodreliesheavilyinthecontributionoftheprivatesector.
Therearetwounderlyingmotivesbehinddatasharingforprivatecompanies.
Firstly,DataPhilanthropywhichistheunderstandingoftheimportanceofdataforsocialgoodandcomplieswithdata-drivenCorporateSocialResponsibilityandsecondly,ensuringthatdevelopingcountriespopulationwillnotreturntopovertylevelswhichdonotallowforviableconsumptionpatterns.
Lastbutnotleast,thepatternofprivacyconcernsisthenorminalmostanyuseoflargedatasetsofuser'sinformation.
EnsuringthattheidentityoftheusercannotbeidentifiedduetohistoricalandlocationdataandthatthedatasetsremainanonymizedwillassistinleveragingBigDatatopreventanytypeofdiscriminationatworkplaceandremainingonpathtoachieveSustainableDevelopment.
CONCLUSIONS:Tomarkthedevelopmentofthenext15years,achievinggenderequalityandempowermentofallwomenandgirlsisthewayforwardfortheregion.
Genderbaseddiscriminationresultsinfemaleworkersbeingdemotedanddismissed.
Discriminationatworkplacefurtherexacerbatesalienationandviolateshumanandlaborrights.
Drawinginformationfromrealtimedatacanassistgovernmentsandtheinternationalcommunitytounderstandfurtherdriversofdiscriminationintheworkplaceandpreventincidentsfromoccurring,ensuringadecentandequitableworkenvironment.
PreparedbyErifyliNomikou,Consultant,EDD/ESCAPReferences:[22,23,24,25,26]UNGlobalPulse,'FeasibilityStudy:IdentifyingtrendsinDiscriminationagainstWomenintheWorkplaceinSocialMedia",GlobalPulseProjectSeriesno11,2014'.
26UNGlobalPulse,'FeasibilityStudy:IdentifyingtrendsinDiscriminationagainstWomenintheWorkplaceinSocialMedia",GlobalPulseProjectSeriesno11,2014'51[27]HowtheUNlabinIndonesiausestwitterAccessibleat:http://www.
fastcolabs.
com/3007178/open-company/how-uns-new-data-lab-indonesia-uses-twitter-preempt-disaster[28]ILO.
DiscriminationatWorkinAsia.
Accessibleat:http://www.
ilo.
org/wcmsp5/groups/public/---ed_norm/---declaration/documents/publication/wcms_decl_fs_89_en.
pdf5.
CaseStudyVUSINGSOCIALMEDIATOMEASUREPUBLICAWARENESSFORCLIMATECHANGEThe2014ClimateSummitandtheupcomingCOP21haveofferedauniqueopportunitytoexploreandmonitorreal-timesocialmediaconversationsaboutclimatechange.
SinceApril2014,UNGlobalPulsehasbeenmeasuringthetotalvolumeoftweets,linksandhashtagsaboutdifferentclimatechangetopicsprovidinginsightonpublicawarenessandengagement.
PROBLEM:Asia-Pacificisoneofregionsmostpronetoclimatechange.
Theimpactsofclimatechangeareprojectedtointensifyinthefutureandworldleadershavepledgedtocombatclimatechangefocusingonadaptionandmitigationtoachievethe2030AgendaforSustainableDevelopment(SDG13).
Climatechangeisnotonlyaregionalprioritybutfortheregiontobesuccessfulinadaptationandmitigation,leadersmustmobilizepeopleandenhancepublicinterestaroundclimatechangeissues.
Therearenotenoughdataavailablethroughtraditionaldatacollectiontoolstoprovideuswithinsightonthepublic'sawarenessandengagementintacklingtheclimatechangechallenge27.
USINGBIGDATATOENACTPOLICIESFORCLIMATECHANGE:Theincreasingnumberofmobileusersinthedevelopingworldandthenewtechnologieswhichofferanunprecedentedopportunityforinterconnectivityandcivicengagementcanbeavaluablesourceofdigitaldataforsocialgood.
Socialmediahasrevolutionizedthewaycitizensrespondtocross-cuttingissues.
Digitaldataisaninnovativewaytogaininsightincitizen'sbehaviorandpromoteparticipatoryandinclusivepolicymaking.
Leveragingtweetsallowcitizenstoundertakeanactiverolethroughsocialmediaoutlets,creatinghighlevelsofcivicengagementandsocialactivismamongthem.
Citizen'sinclusioninclimatechangepoliciesthroughtheironlinepresenceandactiveengagementcanbeagamechangertotheformulationofrespectiveregionalsprioritiesmovingforwardthetransitiontowardssustainabledevelopmentinthepost-2020climateregime.
27Picture:TweetshashtagsthattrendedinrelationtoclimatechangeinFebruary2015,UNGlobalPulsehttp://unglobalpulse.
net/climate/google/52KEYPLAYERS:PolicyMakers,Citizens,UNGlobalPulse,TwitterINSIGHTS&OUTCOMES:Leveragingtweetsthroughthemonitoringofthevolumeandcontentcaninformdecisionandpolicymakersonwhatcitizensaremostlyconcernedaboutandtodevelopcommunicationstotargetpriorityregions.
GlobalPulseandtheSecretaryGeneral'sClimateChangeSupportTeamcreatedatooltomonitorreal-timesocialmediaengagementpriortoandaftertheClimateSummitin2014.
OnadailybasestweetsinEnglish,FrenchandSpanishweremonitoredacrossdifferenttopicsrelatedtoclimatechange.
Measuringandvisualizingtweetsovertimecreatedabaselineofengagement;increasedengagementaroundClimateSummit28.
Hashtags,links,andtweetswereaninnovativeandunprecedentedtooltomeasurepublicengagement,reflectpublicopinionandenactdata-drivenpolicymakingforclimatechange.
Themethodologyusedwasthedevelopmentofataxonomyof1,000wordsandphraseswhichfilteredover15milliontweetssinceApril2014inEnglish,FrenchandSpanish.
Outofeighttopicsthe"economy"and"politics"showedthehighestnumberofpublicconversationsaboutclimatechange.
Thebaselinevolumeremainedat140,000Englishlanguagetweetsperdaywiththatnumberincreasingatover400,000onthedayofeventssuchastheClimateSummitorthePeople'sClimateMarch29.
Followingthesummitthebaselineincreasedbetween10and1530percentindicatingthatclimatechangewasnotatemporaryengagementbutpeoplesustainedtheirinterestinclimatechangeissues.
CONCLUSIONS:Therapidgrowingworldeconomyandpopulationoncethreatenedtocollidewiththeplanet'sfiniteresourcesandfragileecosystems31.
Todaythisthreatisaglobalcrisis.
Climatechangeisacross-cuttingissueandactionisneededimmediately.
Theyear2015iscriticalforsettingtheagendaofthenext15yearsanditcannotbeachievedwithoutactivecitizen'sengagement.
Theabilitytomonitorreal-timeconversationsinsocialmediaanddrawinsightscanbeadrivertomeasureandincreasepublicawarenessandhelpclimatepolicymakerstomakeinformeddecisionsrelevanttotheclimatechangepolicyprioritiesidentifiedbythepeopleineachregion.
Peoplecanbegamechangersisbuildingclimatechangesolutionsforadaptionandmitigation.
PreparedbyErifyliNomikou,Consultant,EDD/ESCAPReferences:[27]UNGlobalPulse,Picture:TweetshashtagsthattrendedinrelationtoclimatechangeinFebruary2015.
Accessibleat:http://unglobalpulse.
net/climate/google/[28,29,30]UNGlobalPulse,"UsingTwittertoMeasureGlobalEngagementonClimateChange",GlobalPulseProjectSeriesno7,2015.
[31]Sachs,J.
"TheAgeofSustainableDevelopment".
ColumbiaPress.
NewYork(2015)28UNGlobalPulse,"UsingTwittertoMeasureGlobalEngagementonClimateChange",GlobalPulseProjectSeriesno7,2015.
29UNGlobalPulse,"UsingTwittertoMeasureGlobalEngagementonClimateChange",GlobalPulseProjectSeriesno7,2015.
30UNGlobalPulse,"UsingTwittertoMeasureGlobalEngagementonClimateChange",GlobalPulseProjectSeriesno7,2015.
31Sachs,J.
"TheAgeofSustainableDevelopment".
ColumbiaPress.
NewYork(2015)53Annex3:Expertcomments,inputs,andrecommendationstoESCAPDraftReportonBigDataandthe2030AgendaforSustainableDevelopmentDuringthecourseoftheESCAPorganizedBigDataandthe2030Agenda:AchievingSustainableDevelopmentintheAsiaandthePacificRegionmeetingheldfrom14-15December2015attheUnitedNationsConferenceCentreinBangkok,Thailand,asessionwasdedicatedtoreceivingcomments,inputs,andrecommendationsonthedraftESCAPreport'Bigdataandthe2030AgendaforSustainableDevelopment',whichwascirculatedforreviewpriortothemeeting.
ThissessionwasfacilitatedbyDr.
AbbasMaaroof,BigDataConsultant,ESCAPandMs.
HalaRazian,AssociateEconomicAffairsOfficer,EnvironmentandDevelopmentPolicySection,EnvironmentandDevelopmentDivision,ESCAP.
TheExpertConsultationonBigDataandthe2030AgendaforSustainableDevelopmentwasattendedby56expertsfromgovernment,civilsociety,theprivatesector,academia,andinternationalorganizations,andESCAPregionalandsubregionalofficesamongothers.
Overall,theexpertsfeltthatthekeyfindingsinthereportwerevalid,andproposedseveralareasoffurtherresearchandelaboration.
Thecomments,inputs,andrecommendationsreceivedareincludedbelow:Summaryofcommentsandrecommendations:OversimplifiedconceptofBigData.
ItwasrecommendedthatthedefinitionofBigDatabebroadenedtobeyonda'bigdataset',andtoincorporatehowBigDataislinkedtoothersourcesofdataandthebroaderelementsofthedataecosystem.
ThisincludesconsiderationofhowgovernmentsproduceanduseBigData,aswellasefficienciesintermsofeconomicvalue.
BigDatashouldnotbeelevatedaboveotherdataSources.
ThereportshouldmakerecommendationsonwhatweneedtounderstandtobringBigDataforward,inthecontextofotherdatasourcesthatarealreadyreadilyavailableandinthecontextofthedatarevolution.
ItwouldbeinterestingtolookatspecificSDGs,whatdataisavailable,andwheregapsexistthatcouldbefilledwithBigData,allthewhileacknowledgingtherisksinvolvedandthecomplementarysourcesthatneedtobeintegrated.
IncreasedfocusonSDGs,thethreedimensionsofsustainability,andtheirindicators,wouldbebeneficial.
AdditionalinformationonhowBigDatacancrossboundariesacrossthethreedimensionsofsustainabledevelopmentwouldbeuseful.
ThiscouldincludeprovidingconcreteexamplesofhowBigDatahasbeenusedtosupportimplementation,followupandreviewofanindividualSDG,including:investigatingthesystemdynamicsbetweentheSDGtargetsandindicators;lookingatthenexusbetween54differentSDGsandhowbigdatacouldbeusedtosupportthiskindofinterconnectionforimplementation,evaluationandintegratedpolicymaking.
DrilldownbeyondthenationalleveltotheusesofBigDataatsub-nationalandlocallevels.
ItwasrecommendedthatexamplesofhowBigDatawasbeingusedatsubnationallevelsbefurtherelevatedinthereport.
Identifycasestudiesthataregovernmentdriven.
Whilethecasestudiesidentifiedareuseful,itwouldbeequallybeneficialtoidentifygovernment-ledcasestudies.
Increasedfocusontheissueofprivacy.
Abetterunderstandingofthemeaningofprivacycouldbeputforwardinthereport,includingaunifiedlegalandethicalapproach,howtomoveforward,andhowtotacklethechallengenotonlyatnationalbutalsoatglobalandregionallevels.
Increasedstratificationofexamples.
Thereportcoulddemonstrateagreaterappreciationofdifferencesbetweenhighlydevelopedeconomieswithskilledandknowledgeablepublicserviceorganizations;versusothercountrieswhere'datafication'isasyetlow.
Differencesexistbetweendatausersandtheirsituationsandneeds,andthisshouldbemademoreexplicit.
Increasedstratificationofstakeholders.
Universitiesshouldplayabiggerroleinthereportasleadingstakeholders.
Theycouldserveasaforumforconversation,andasneutralyetactiveparticipantsinBigDataanalysis,andasameansofcapacitybuilding.
Conceptofdataphilanthropytobecriticallyexamined.
DataPhilanthropyisacontestedterm.
Someparticipantsvoicedthatitshouldratherbegovernmentsmandating,orregulating,privatesectordatadisclosure.
WhileDataPhilanthropyisanimportantaspectoftheBigDataecosystem,Governmentsroleinsettingtheframeworkfordisclosureandopendatashouldalsobeconsidered.