temporal graph analytics with gradoop
play

Temporal Graph Analytics with GRADOOP Christopher Rost and Kevin - PowerPoint PPT Presentation

Temporal Graph Analytics with GRADOOP Christopher Rost and Kevin Gomez Leipzig University About the speakers and the team Kevin, Christopher, Prof. Dr. Erhard Rahm, PhD Student PhD Student Database Chair Timo, Lucas, M.Sc. Student


  1. Temporal Graph Analytics with GRADOOP Christopher Rost and Kevin Gomez Leipzig University

  2. About the speakers and the team Kevin, Christopher, Prof. Dr. Erhard Rahm, PhD Student PhD Student Database Chair Timo, Lucas, M.Sc. Student B.Sc. Student

  3. We’re hiring! CENTER FOR SCALABLE DATA ANALYTICS AND ARTIFICIAL INTELLIGENCE ● Visit us at www.scads.de ● Open positions and projects (Developer and PhD Students) ● Various topics: ● Machine Learning (Graph-ML, Privacy) ● Image Recognition ● Blockchain (Distributed-Ledger) ● Graph Analytics

  4. Motivation

  5. “Graphs are everywhere”

  6. “Graphs are everywhere”

  7. “Graphs are everywhere” Mike Raj Sheldon O f d n i e f r f r John i e n d f O d friendOf O n e f i r f Bob friendOf friendOf James Anna friendOf friendOf friendOf Dutch Chris friendOf friendOf Sally friendOf Arthur friendOf Jack

  8. “Graphs are everywhere” Mike Raj Sheldon O f d e n r i f f r John i e n age:29 d f O d friendOf O sex: f n e f i r f yob: 1991 Bob friendOf friendOf James Anna friendOf friendOf friendOf Dutch Chris friendOf friendOf Sally friendOf Arthur friendOf Jack

  9. “Graphs are everywhere” Mike Raj Sheldon O f d n i e f r f r John i e n age:29 d f O d friendOf O sex: f n e i f r f yob: 1991 Bob friendOf friendOf James Anna friendOf since: 5Y friendOf friendOf Dutch Chris friendOf friendOf Sally friendOf Arthur friendOf Jack

  10. “A open-source framework and research platform for efficient , distributed and domain independent management and analytics of heterogeneous and temporal graph data.“

  11. Architecture Java 8 G raph A nalytical L anguage (GrALa) 45k LOC TPGM EPGM ALv 2.0 Distributed Operator Implementation HDFS / YARN

  12. Graphs and collections of graphs

  13. Operators Operators Algorithms Binary Unary Logical Graph Frequent Subgraph Graph Collection

  14. Temporal extension

  15. Temporal extension

  16. Temporal extension

  17. Temporal extension

  18. Time dependent operators Operators Algorithms Binary Unary Logical Graph Frequent Subgraph Graph Collection

  19. Time dependent operators Operators Algorithms Binary Unary (Temp-)Logical Graph Frequent Subgraph (Temp-)Graph Collection

  20. Use case: spread of airborne pathogens Accident Oncology Virus X Symptoms: 3 rd eye growing Transmission: if contact > 5 min Incubation period: 5 time units Surgery Cardiology In case of an infection, which hospital services are Contact tracking Sensors capturing who is close to whom at risk of contracting at what time. the virus X? {empID1, empID2, t_from, t_to} Images by Vecteezy.com.

  21. TemporalGraph contacts = TemporalGraph contacts = mySource.getTemporalGraph(); mySource.getTemporalGraph(); t 0 t 1 t 5 ... t 10 t 15 t 20 time

  22. contact [ t 2 ,t 6 ) Employee [-,-) id: 32432 srv: Oncology Employee [-,-) id: 99934 srv: Oncology t 0 t 1 t 5 ... t 10 t 15 t 20 time

  23. Breaking news: Employee of oncology infected by VirusX. Detected at t 20 . t 0 t 1 t 5 ... t 10 t 15 t 20 time

  24. Breaking news: Employee of oncology infected by VirusX. Detected at t 20 . Virus X Symptoms: 3 rd eye growing Transmission: if contact > 5 min Incubation period: 5 time units t 0 t 1 t 5 ... t 10 t 15 t 20 time

  25. TemporalGraph contacts = TemporalGraph contacts = mySource.getTemporalGraph(); mySource.getTemporalGraph(); contacts = contacts.snapshot snapshot( contacts = contacts.snapshot snapshot( new FromTo(t15, t20)); new FromTo(t15, t20)); Virus X Symptoms: 3 rd eye growing Transmission: if contact > 5 min Incubation period: 5 time units t 0 t 1 t 5 ... t 10 t 15 t 20 time

  26. TemporalGraph contacts = TemporalGraph contacts = mySource.getTemporalGraph(); mySource.getTemporalGraph(); contacts = contacts.snapshot snapshot( contacts = contacts.snapshot snapshot( new FromTo(t15, t20)); new FromTo(t15, t20)); Virus X Symptoms: 3 rd eye growing Transmission: if contact > 5 min Incubation period: 5 time units t 0 t 1 t 5 ... t 10 t 15 t 20 time

  27. TemporalGraph contacts = TemporalGraph contacts = mySource.getTemporalGraph(); mySource.getTemporalGraph(); contacts = contacts.snapshot( contacts = contacts.snapshot( new FromTo(t15, t20)); new FromTo(t15, t20)); contacts = contacts.groupBy groupBy( contacts = contacts.groupBy groupBy( (v v[‘srv’]), � (v v[‘srv’]), � [], [], (e e.getLabel()), � (e e.getLabel()), � [new MaxDuration()]); [new MaxDuration()]); Virus X Symptoms: 3 rd eye growing Transmission: if contact > 5 min Incubation period: 5 time units t 0 t 1 t 5 ... t 10 t 15 t 20 time

  28. TemporalGraph contacts = TemporalGraph contacts = mySource.getTemporalGraph(); contact mySource.getTemporalGraph(); maxDur: 124 min contact maxDur: 40 min contacts = contacts.snapshot( contacts = contacts.snapshot( new FromTo(t15, t20)); new FromTo(t15, t20)); contact maxDur: 9 min contacts = contacts.groupBy groupBy( contacts = contacts.groupBy groupBy( (v v[‘srv’]), � (v v[‘srv’]), � [], contact [], maxDur: 122 min (e e.getLabel()), � (e e.getLabel()), � [new MaxDuration()]); contact [new MaxDuration()]); maxDur: 2 min contact maxDur: 146 min Virus X Symptoms: 3 rd eye growing Transmission: if contact > 5 min Incubation period: 5 time units t 0 t 1 t 5 ... t 10 t 15 t 20 time

  29. TemporalGraph contacts = TemporalGraph contacts = mySource.getTemporalGraph(); contact mySource.getTemporalGraph(); maxDur: 124 min contact maxDur: 40 min contacts = contacts.snapshot( contacts = contacts.snapshot( new FromTo(t15, t20)); new FromTo(t15, t20)); contact maxDur: 9 min contacts = contacts.groupBy groupBy( contacts = contacts.groupBy groupBy( (v v[‘srv’]), � (v v[‘srv’]), � [], contact [], maxDur: 122 min (e e.getLabel()), � (e e.getLabel()), � [new MaxDuration()]); contact [new MaxDuration()]); maxDur: 2 min contact maxDur: 146 min Virus X Symptoms: 3 rd eye growing Transmission: if contact > 5 min Incubation period: 5 time units t 0 t 1 t 5 ... t 10 t 15 t 20 time

  30. TemporalGraph contacts = TemporalGraph contacts = mySource.getTemporalGraph(); contact mySource.getTemporalGraph(); maxDur: 124 min contact maxDur: 40 min contacts = contacts.snapshot( contacts = contacts.snapshot( new FromTo(t15, t20)); new FromTo(t15, t20)); contact maxDur: 9 min contacts = contacts.groupBy groupBy( contacts = contacts.groupBy groupBy( (v v[‘srv’]), � (v v[‘srv’]), � [], contact [], maxDur: 122 min (e e.getLabel()), � (e e.getLabel()), � contact [new MaxDuration()]); [new MaxDuration()]); maxDur: 2 min Breaking news: contact maxDur: 146 min Employees of oncology and surgery quarantined because of VirusX. t 0 t 1 t 5 ... t 10 t 15 t 20 time

  31. Conclusion ● Distributed graph analysis system ● Temporal property graph model – Bitemporal support – Logical graphs and graph collections – Composing operators and algorithms ● Declarative workflow creation ● Visit Gradoop: http://gradoop.com ● Read our Wiki: https://github.com/dbs-leipzig/gradoop/wiki ● Start: https://github.com/dbs-leipzig/gradoop/wiki/Getting-started ● Try the examples: https://github.com/dbs-leipzig/gradoop/wiki/Examples

Recommend


More recommend