WIXLIB

Master Degree inUniversity of Milano-Bicocca1

Departments involved Dipartimento di Informatica, Sistemistica eComunicazione (DISCo) Dipartimento di Economia, Metodi quantitativi eStrategie di impresa Dipartimento di Statistica e Metodi quantitativi2

The three stakeholders, students, companies,teachers: how to boost cooperation among them?Coursesand LabsEarly BirdInitiativeKaggle3

Students4

Statistics on enrolled Students - 1Per area culturaleEconomia e caMatematicaScienze ComFilosofia0,90,90,90,95,57,338,516,519,35

Statistics on enrolled Students - 2Per area geografica di provenienza1,19,97,714,355,45,525,3BicoccaAltre MilanoLombardiaaltre NordCentroSud6

The three stakeholders, students, companies,teachers: how to boost cooperation among them?Coursesand Labs7

From S. Ceri, EDBT Venice, March 2017How big is the genome?As a string: 700MByteAs raw data: 200 GbyteAs called mutations: 125MByteHow many genomes will be sequencedin 5 years?Estimates: order of 5-20 MillionsVery big data problem

From small data to big dataBroadness of observed realityTimeDepth in knowledge of observed reality9

10

From small data to big dataBroadness of observed realityTimeDepth in knowledge of observed reality11

1 among 4Courses1 among 3CYB – Cybersecurityfor data scienceIP – Signal and imageprocessing1 among 3DM&DV - Datamanagementand visualizationMLDM – MachineLearning &Decision ModelsJSI – Juridical &Social Issues inInformationSocietySTDA – StatisticalmodellingTIDS – Technological infrastructures for data scienceFC – Foundam.in Comp.Sc.1 among 3DS – DataSemanticsIS – InformationSystems1 among 2FS – Found.in Stat. & P.WM&CM – Webmarketing &CommunicationManagementTMS – Textminingand searchEW – Expert WeekData Science Lab inEnvironment & PhysicsBDGIS – Big Data in Geographical Information SystemsBDPhis - Big data managementand analysis in physics researchData Science Lab in biosciencesBDB&B – Big data inbiotechnology & biosciencesMSBD - Making sense ofbiological dataData Science Lab in MedicineASM – Advanced statisticalmethodologies for Big DataBDM1 – - Big Datain Health CareSDM – Streamingdata management andtime series analysisBDM2 - Medical imaging& big dataEDS – Economics for DataScience1 among 3SMA – Social MediaAnalyticsSS –Service ScienceDSL - Data ScienceLabBI - BusinessIntelligenceLabsIL – Industry LabData Science Lab in Business & MarketingBDBF – - Big data in Businessand FinanceBDBP - Big data inBehavioural PsycologyData Science Lab in Public Policies & ServicesBDPHe – - Big Datain Public HealthBDPS - Big Data in Public andSocial ServicesFirst year1 among 3Second year12

Courses1 among 41 among 3Common coursesCYB – Cybersecurityfor data scienceAnalytical trackBusiness trackAnalyticaltrack1 among 3DM&DV - Datamanagementand visualizationMLDM – MachineLearning &Decision ModelsJSI – Juridical &Social Issues inInformationSocietySTDA – StatisticalmodellingFC – Foundam.in Comp.Sc.IP – Signal and imageprocessingTIDS – Technological infrastructures for data science1 among 3DS – DataSemanticsIS – InformationSystems1 among 2FS – Found.in Stat. & P.WM&CM – Webmarketing &CommunicationManagementDSL - Data ScienceLabTMS – Textminingand searchBDPhis - Big data managementand analysis in physics researchData Science Lab in biosciencesBDB&B – Big data inbiotechnology & biosciencesMSBD - Making sense ofbiological dataData Science Lab in MedicineBDM1 – - Big Datain Health CareSDM – Streamingdata management andtime series analysisBDM2 - Medical imaging& big data1 among 3BusinessTrackBDGIS – Big Data in Geographical Information SystemsASM – Advanced statisticalmethodologies for Big DataEDS – Economics for DataScienceEW – Expert WeekData Science Lab inEnvironment & PhysicsSMA – Social MediaAnalyticsSS –Service ScienceBI - BusinessIntelligenceLabsIL – Industry LabData Science Lab in Business & MarketingBDBF – - Big data in Businessand FinanceBDBP - Big data inBehavioural PsycologyData Science Lab in Public Policies & ServicesBDPHe – - Big Datain Public HealthBDPS - Big Data in Public andSocial ServicesFirst year1 among 3Second year13

Scientific areas1 among 41 among 3Computer ScienceCYB – Cybersecurityfor data scienceStatisticsSocioEconomicMixed1 among 3DM&DV - Datamanagementand visualizationAnalyticaltrackASM – Advancedstatistical methodologiesfor Big DataDS – DataSemanticsIS – InformationSystemsJSI – Juridical &Social Issues inInformationSociety1 among 2FS – Found.in Stat. & Pr.STDA – StatisticalmodellingWM&CM – Webmarketing &CommunicationManagementTIDS – Technological infrastructures for data science1 among 3FC – Foundam.in Comp.Sc.MLDM – MachineLearning &Decision ModelsIP – Signal and imageprocessingSDM – Streamingdata management andtime series analysisTMS – Text miningand searchEDS – Economics for DataScience1 among 3EW – Expert WeekBusinessTrackSMA – Social MediaAnalyticsSS –Service ScienceBI - BusinessIntelligenceDSL - Data Science LabData Science Lab inEnvironment & PhysicsBDGIS – Big Data in Geographical Information SystemsBDPhis - Big data managementand analysis in physics researchData Science Lab in biosciencesBDB&B – Big data inbiotechnology & biosciencesMSBD - Making sense ofbiological dataData Science Lab in MedicineBDM1 – - Big Datain Health CareBDM2 - Medical imaging& big dataLabsIL – Industry LabData Science Lab in Business & MarketingBDBF – - Big data in Businessand FinanceBDBP - Big data inBehavioural PsycologyData Science Lab in Public Policies & ServicesBDPHe – - Big Datain Public HealthBDPS - Big Data in Public andSocial ServicesFirst year1 among 3Second year14

DSc – a dynamic, evolving science1 among 3StatisticsCYB – Cybersecurityfor data scienceComputer ScienceSocioEconomicMixed1 among 3DM&DV - Datamanagementand visualizationAnalyticaltrackASM – Advancedstatistical methodologiesfor Big DataDS – DataSemanticsIS – InformationSystemsJSI – Juridical &Social Issues inInformationSociety1 among 2FS – Found.in Stat. & P.STDA – StatisticalmodellingWM&CM – Webmarketing &CommunicationManagementDSL - Data Science LabTIDS – Technological infrastructures for data science1 among 3FC – Foundam.in Comp.Sc.MLDM – MachineLearning &Decision ModelsIP – Signal and imageprocessingSDM – Streamingdata management andtime series analysisTMS – Text miningand searchEDS – Economics for DataScience1 among 3EW – Expert WeekSMA – Social MediaAnalyticsBusinessTrackSS –Service ScienceBI - BusinessIntelligence1 among 4Data Science Lab inEnvironment & PhysicsBDGIS – Big Data in Geographical Information SystemsBDPhis - Big data managementand analysis in physics researchData Science Lab in biosciencesBDB&B – Big data inbiotechnology & biosciencesMSBD - Making sense ofbiological dataData Science Lab in MedicineBDM1 – - Big Datain Health CareBDM2 - Medical imaging& big dataLabsIL – Industry LabData Science Lab in Business & MarketingBDBF – - Big data in Businessand FinanceBDBP - Big data inBehavioural PsycologyData Science Lab in Public Policies & ServicesBDPHe – - Big Datain Public HealthBDPS - Big Data in Public andSocial ServicesFirst year1 among 3Second year15

Four V’s of Big Data Volume Velocity Variety Value16

Change of Paradigm in Data Management SystemsVolumeBig DataNoSQL Hadoop MapReduceHadoop & Spark(plus: distributed file system)SmallDataSQL Traditional DBMSLong-termchanging dataSpark(plus: in-memory processing)StreamingdataVelocity17

The four Vs:1. VOLume2. VELocity3. VARiety4. VALue1 among 41 among 3DM&DV - Datamanagementand visualizationVALMLDM – MachineLearning &Decision ModelsVALJSI – Juridical &Social Issues inInformationSocietySTDA – StatisticalmodellingDSL1 – DataScience Lab 1VELCYB – Cybersecurityfor data scienceAnalyticaltrack1 among 3VOLData Science Lab inEnvironment & PhysicsFC – Foundam.in Comp.Sc.DS – DataSemanticsIP – Signal and imageprocessingVELTIDS – Technological infrastructures for data science1 among 3VOLASM – Advanced statisticalmethodologies for Big DataVALIS – InformationSystems1 among 2FS – Found.in Stat. & PrSDM – Streamingdata management andtime series analysisVOLTMS – Textminingand search VELVALWM&CM – Webmarketing &CommunicationManagementVOLVELVALEDS – Economics for DataScienceEW – Expert Week1 among 3BusinessTrackSMA – Social MediaAnalyticsSS –Service ScienceBI - BusinessIntelligenceData Science LabBDPhis - Big data managementand analysis in physics researchData Science Lab in biosciencesVOLVARBDGIS – Big Data in Geographical Information SystemsVALBDB&B – Big data inbiotechnology & biosciencesMSBD - Making sense ofbiological dataData Science Lab in MedicineBDM1 – - Big Datain Health CareBDM2 - Medical imaging& big dataIL – Industry LabData Science Lab in Business & MarketingBDBF – - Big data in Businessand FinanceBDBP - Big data inBehavioural PsycologyData Science Lab in Public Policies & ServicesBDPS - Big Data in Public andSocial Services1 among 3Second yearVELVALBDPHe – - Big Datain Public HealthFirst yearVOLVAL

Data types Tables Relational (keys, referential integrity, etc.) Weak semantics (e.g. csv) Texts Loosely structured Semistructured (e.g. XML) Signals (from the Internet of Things) Images (e.g. X-ray, security, etc.) Graphs Mathematical - syntactic Knowledge - semantic Open data & Linked Open Data Maps & Remote sensing & Georeferenced data Mixed (Web data)20

Main Data Types1 among 41 among 3Tables & SeriesSignals and imagesKnowledge graphsLoosely Str.&Semistr. texts AnalyticalMaps & Georef. datatrackNot relevantDM&DV - Datamanagementand visualizationMLDM – MachineLearning &Decision ModelsJSI – Juridical &Social Issues inInformationSocietySTDA – Statisticalmodelling1 among 3FS – Foundam.in InformaticsDS – DataSemanticsIS – InformationSystems1 among 2FS – Found.in Stat. & PrWM&CM – Webmarketing &CommunicationManagementCYB – Cybersecurityfor data scienceBDGIS – Big Data in Geographical Information SystemsIP – Signal and imageprocessingBDPhis - Big data managementand analysis in physics researchIP – Signal and imageprocessingData Science Lab in biosciencesTIDS – Technological infrastructures for data science1 among 3MSBD - Making sense ofbiological dataData Science Lab in MedicineBDM1 – - Big Datain Health CareSDM – Streamingdata management andtime series analysisBDM2 - Medical imaging& big dataEDS – Economics for DataScienceEW – Expert WeekSMA – Social MediaAnalytics1 among 3SS –Service ScienceBI - BusinessIntelligenceDSL1 – DataScience Lab 1First yearBDB&B – Big data inbiotechnology & biosciencesASM – Advanced statisticalmethodologies for Big DataTMS – Textminingand searchBusinessTrackData Science Lab inEnvironment & PhysicsLabsIL – Industry LabData Science Lab in Business & MarketingBDBF – - Big data in Businessand FinanceBDBP - Big data inBehavioural PsycologyData Science Lab in Public Policies & ServicesBDPHe – - Big Datain Public HealthBDPS - Big Data in Public andSocial ServicesDSL1 – DataScience Lab 11 among 3Second year21

Traditional Analysis Life cycle vsnew Analysis life cycle of digital (big) dataBig DataLife cycleTraditionallife tionAnalysisDiffusionCross cutting activitiesSEMANTICSQUALITYLEARNINGVALUE22

Phases of the life cycle and main feedbacks1. Access2. Management3. Visualization4. Analysis5. Diffusion23

Phases of the life cycle - detail1.Access & Acquisition Search Selection Acquisition2. Management Filtering Quality assessment Semantic interpretation & enrichment Matching & integration3. Visualization4. Analysis Descriptive analysis: what happened or what is happening. Diagnostic analysis: why it happened or why it is happening. Predictive analysis: what will happen Prescriptive analysis : what to do to achieve the goal5. Diffusion24