Page 1
Ravi VatrapuDirector, Centre for Business Data Analytics (bda.cbs.dk)
Professor, Department of IT ManagementCopenhagen Business School, Denmark
Email: [email protected] : http://www.cbs.dk/en/staff/rvitm
Centre: http://bda.cbs.dk
Transforming Big Data Sets into Business Assets
Page 2
2
• Phenomena• Internet, Social Media & Society• Challenges & Opportunties: In-House Data + Big Data• Business Value: Big Data Sets à Business Assets
• Centre for Business Data Analytics (bda.cbs.dk)• Meaningful Facts• Actionable Insights• Valuable Outcomes• Sustainable Impacts
• Case Projects• Predictive Models• Prescriptive Analytics• Visual Analytics
• Our Product and Service Portfolio
Outline
Page 3
3
About Me: Global Nomad
Vizag,India Blacksburg,USA
Honolulu,USACopenhagen,Denmark
Page 4
4
Part I:Phenomena
Page 5
5
Internet, Social Media & Society
https://en.wikipedia.org/wiki/On_the_Internet,_nobody_knows_you're_a_dogRamu:“OntheFacebook,everybodyknowsIamadog"
Page 6
6
Challenges: How to Combine House Data with Big Data
Page 7
7
Opportunities: Business Value = In-House Data + Big Data
Porta,M.,House,B.,Buckley,L.&Blitz,A.(2008)
Page 8
8
Business Value = In-House Data + Big Data
Wollan,R.,Smith,N.&Zhou,C(2011)SonyPS4Controller:“Share”Button
ImagefromKotaku
Page 9
9
Big Data Sets à Business AssetsCase: Product: Baby-Monitors
MasterThesis:AdeleIndianeGurrich Kristensen&StineSofieBragdø
Page 10
10
Part II: CSSL ApproachSet-Theoretical Big Social Data Analytics
Page 11
11
CentreforBusinessDataAnalytics(cbsBDA)locatedattheDept.ofITManagement,CopenhagenBusinessSchool.cbsBDA conductstransdisciplinarybasicresearchonsocio-technicalinteractions withspecificapplicationstomanagersincompanies,teachersinschoolsandresidentsincities.
1Director&Professor2AssistantProfessors10PhDStudents4ResearchAssociates+11FacultyCollaboratorsatCBS,KUandbeyond
(IEEEEDOC2014)
cbsBDA (bda.cbs.dk)
Page 12
12
cbsBDA’s Naïve Model for Applied Research
• Symptoms
• Diagnosis
• Therapy• Prescription• Proscription
• Prognosis• Positive/Negative
MARCELLUS:
SomethingisrotteninthestateofDenmark
http://shakespeare.mit.edu/hamlet/full.html
https://en.wikipedia.org/wiki/File:Helsing%C3%B8r_Elsinore_from_sea_01.jpg
Page 13
13
Class of Problems: Social Associations (Organisations)
Page 14
14
SocialData
Interactions Conversations
Actors ArtifactsActivitiesActions Topics EmotionsPronounsKeywords
Source:RaviVatrapu
Conceptual Model of Social Data
Page 15
15
Analytical Framework for Set-Theoretical CSS
Page 16
McDonaldsDKActors:266,000Noma Actors:4,567McDonaldsDK&Noma Actors:203
CROSS-WALL ANALYSIS:MCDONALDS DKVS.NOMA
9
• Whatkindsofsocialtextdothese203 actorscreate,circulate,andinteractwith?
• What,ifany,isthecross-culturalvariationofactorsassociatingwithbothfastfoodandfinedining?
Page 17
17
Part III: Case Projects
Page 18
18
Case Project #1: Loyalty Club Programs
Page 19
19
• Datasets• CRM
• Interviews
• Facebook
Case Project #1: Loyalty Club Programs
Page 20
© Temperaturenpådanskeloyalitetsklubberanno2015-II 20
BigSocialDataAnalytics
Page 21
21
Foragivensocialmediaaction,wewanttoanalyse andmodel:
• UserCharacteristics• Emotion• Personality
• User/ConsumerCharacteristics• ConsumerDecision-MakingStage
• Organisational Consequences• BrandSentiment
• SocialMediaConsequences• SocialEngagementPotential
Beyond Social Media-->Towards Social Business
Page 22
BeyondSocialMedia-->TowardsSocialBusiness“Heres anidea.Ifyouliketheirfoodeatthere.Ifyoudont liketheirfoodeatsomewhereelseormakeyourownmeal.
Ireallydont understandwhatthebigdealis.”
User Consumer
Organisation
SocialInfluence
Page 23
TextClassification:Multi-DimensionalModels
Page 24
BasicEmotions
0.00% 20.00% 40.00% 60.00% 80.00% 100.00%
BRmatascoop
ForbrugsforeningenIKEA
imercolOplus
Sportsmaster
BasicEmotions:Proportion
Joy% Sadness% Surprise% Fear% Disgust% Anger%
0 0.2 0.4 0.6 0.8 1
BRmatascoop
ForbrugsforeningenIKEA
imercolOplus
Sportsmaster
BasicEmotions:Intensity
JoyIntensity SadnessIntensity SurpriseIntensity
FearIntensity DisgustIntensity AngerIntensity
Page 25
BrandParameters:HistoricalDevelopment:UserEmotionsvs.BrandSentiment:
Page 26
PredictingNetPromotorScoreFromBigSocialData
R²=0.95813
0
10
20
30
40
50
60
0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70NPS Poly.(NPS)
Page 27
27
Case Project #2: Market & User Segmentation
Page 28
CROSS-WALL ANALYSIS:USER/CUSTOMER SEGMENTATION
28
DK2011 US2008
Page 29
ENGAGEMENT DIMENSIONS &USER SEGMENTS
29
SOCIAL NETWORK DIALOG SPACE
Robertson,S.,Vatrapu,R.,&Medina,R.(2010).OfftheWallPoliticalDiscourse:FacebookUseinthe2008U.S.Presidential Election.InformationPolity,15(1-2),11-31.
Page 30
BUSINESS VALUE:REAL PROMOTER SCOREProduct Advocates are champions for products in general.
Product Enthusiasts are the users that aspire for the product category.
Brand Loyalists are champions of a particular brand.
Brand Tourists are in the early stages of brand consideration.
Page 31
31
Case Project #3: Sales & Revenue Forecasters
Page 32
32
Business Value: Sales and Revenue Predictive Models
(IEEEEDOC2014)(ICCSS2015)
28
30
32
34
36
38
40
42
44
46
Q1'10
Q2'10
Q3'10
Q4'10
Q1'11
Q2'11
Q3'11
Q4'11
Q1'12
Q2'12
Q3'12
Q4'12
Q1'13
Q2'13
Q3'13
Q4'13
Q1'14
Q2'14
Q3'14
Q4'14
Q1'15
H&Msales,billionSEKperQuarter
Sales PredictedSales
Company DataSource TimePeriod SizeofDatasetApple Twitter 2007® October12,
2014500million+tweetscontaining“iPhone”
H&M Facebook January01,2009®October12,2014
~15millionFacebookevents
Page 33
33
Case Project #4: Social Media Crises (“Shitstorms”)
Page 34
34
CSR Crises: Bangladesh Factory Accidents & Volkswagen
IEEEEDOC2015 IEEEBigData2015
Page 35
35
Business Impact: Social Media Crisis
(ACMCABS2014) (IEEEEDOC2015)(IEEEBigData2015)
DuringCrisis:05-19February,2014Artefacts:AllData:WallbeginningtolastcollectedtimeActors:AllFacebookusersonCopenhagenZooPageActions:LIKE
Activity:PositiveAssociationSociologicalImportanceOrganizationalRelevance
Interpretation:ComputationalSocialScience:SetTheoryLIKEswereawayofexpressingculturalsolidarityandin-groupsupporttoaDanishinstitutionperceivedtobeunderundeservedout-groupcriticism
LikesonZoo’sPosts&CommentsUniqueActorsonZoo’sFBWall
Page 36
36
Case Project #5: EU Immigration Crisis
Page 37
37
EU Immigration Crisis
BDACourseProject:Jensen,Brock,Hody,Christensen&AlHumaidan
Page 38
38
• Big Social Data• Complete Corpus for Facebook• Multi-Channel, Multi-Language & Multi-Domain
• Analytics Software• Social Data Analytic Tool (SODATO)• Social Set Visualiser (SOSEVI)• Multi-Dimensional Text Analytics (MUDITA)• Social Business Predictor (SB-PRE)• Social Business Integrator (SB-INT)
• Research & Consultancy Reports
• Analytics Time Horizons• Fixed• Incremental• Continuous
• Analytics Mode• Historic• Near Real-Time• Real-Time
• Projects• Research
• Research Projects• Industrial PhD Projects
• Consultancy• Real Promoter Score• Strategic Management
Our Product Portfolio
Page 39
39
Interested?Contact us!