Download as PDF, PPTX
Uploaded byNeo4j
Are You Underestimating the Value Within Your Data? A conversation about graph technology
The document discusses the growing importance of graph technology in understanding and analyzing complex relationships within data, emphasizing its effectiveness over traditional data processing methods. It highlights various applications of graph data science, including enhanced analytics, fraud detection, and customer insights, alongside the potential for significant improvements in prediction accuracy. Additionally, the document encourages organizations to adopt graph technology and engage with communities to leverage its capabilities for business innovation.
More Related Content
Similar to Are You Underestimating the Value Within Your Data? A conversation about graph technology
Are You Underestimating the Value Within Your Data? A conversation about graph technology
- 1. © 2022 Neo4j,Inc. All rights reserved. © 2022 Neo4j, Inc. All rights reserved. 1 Are You Underestimating the Value Within Your Data? A conversation about graph technology Dr Jesús Barrasa Head of Field Engineering, Neo4j Dr Jim Webber Chief Scientist, Neo4j
- 2. © 2022 Neo4j,Inc. All rights reserved. © 2022 Neo4j, Inc. All rights reserved. 2 What’s a graph?
- 3. © 2022 Neo4j,Inc. All rights reserved. 3 This is a chart
- 4. © 2022 Neo4j,Inc. All rights reserved. 4 This is a graph
- 5. © 2022 Neo4j,Inc. All rights reserved. © 2022 Neo4j, Inc. All rights reserved. Relationships are the strongest predictor of behavior 5 But you can’t analyze what you can’t see • Legacy data processing techniques ignore relationships • Graphs are built on relationships, so… • You don’t have to guess at the correlations: with graphs, relationships are built in
- 6. © 2022 Neo4j,Inc. All rights reserved. 6 Labeled Property Graph Model Nodes • Can have Labels to classify nodes Relationships • Relate nodes by type and direction • Jim likes soccer, soccer does not like Jim Properties • Stored as name/value pairs Performance • Traversals are always O(1) • Query latency depends on how much of the graph you want to explore • It does not depend on data set size
- 7. © 2022 Neo4j,Inc. All rights reserved. 7 Graphs are everywhere Networks Transport, power, telecoms, data, social Knowledge MDM, Customer 360, Pedagogy, sciences Personalization Recommendations, Disambiguation, patient/customer journey Discovery IT management, Drug discovery, Process innovation, Digital Twin Decision Analysis Cybersec, Fraud, Policing, Intelligence, forecasting
- 8. © 2022 Neo4j,Inc. All rights reserved. 8
- 9. © 2022 Neo4j,Inc. All rights reserved. © 2022 Neo4j, Inc. All rights reserved. 9 Why are graphs everywhere right now?
- 10. © 2022 Neo4j,Inc. All rights reserved. 10 Why now? A timeline 2009 2011 2015 2022
- 11. © 2022 Neo4j,Inc. All rights reserved. © 2022 Neo4j, Inc. All rights reserved. I’ve got lots of data, how do I use analytics to get insights? 11
- 12. © 2022 Neo4j,Inc. All rights reserved. 12 Enhanced Analytics with Graph Data Science Graph Data Science BI & VISUALIZATIONS DHW STORE PROCESS MACHINE LEARNING Cloud Functions Neo4j Bloom DataProc Analytics Feature Engineering Data Exploration Graph Data Science Business Applications & Existing Systems Files (unstructured, structured) TensorFlow KNIME Python Cloud Storage AWS Lambda
- 13. © 2022 Neo4j,Inc. All rights reserved. 13 Graphs Contain Implicit Knowledge Which of the colored nodes would be considered the most ‘important' ?
- 14. © 2022 Neo4j,Inc. All rights reserved. 14 Graph Algorithms Help Unlock This Knowledge D D has the highest valence This is the most connected individual in the network. If being well-known is important, pick D. G has the highest closeness centrality (0.52) Information will disperse through the network more quickly through this individual. If you need to get a message out rapidly, choose them. G I has the highest betweenness centrality (0.59) This person is an efficient connector of other people. Risk of network disruption is higher if you lose this individual. I Most Important?
- 15. © 2022 Neo4j,Inc. All rights reserved. Better Predictions with Data You Already Have ● Traditional ML ignores network structure because it’s difficult to extract ● Graph features enrich existing ML pipelines to increase accuracy, or ● Graphs use relationships to unlock otherwise unattainable predictions 15 Machine Learning Pipeline
- 16. © 2022 Neo4j,Inc. All rights reserved. 16 Machine Learning Pipelines Graph Native ML Pipelines: ● Node Classification ● Node Regression (new!) ● Link Prediction Trained Models & the Model Catalog: ● Save, persist & publish trained models ● Instantly apply for prediction on new and existing data in the graph Guided & Automated Pipeline Features: ● Data splitting & rebalancing ● Feature engineering ● Model evaluation and selection ● Automated hyperparameter tuning (new!) What we do : Link Prediction Example Define a pipeline, add the steps you want to perform, and we handle the rest for you.
- 17. © 2022 Neo4j,Inc. All rights reserved. © 2022 Neo4j, Inc. All rights reserved. 17 “By 2025, Graph technologies will be used in 80% of data and analytics innovations, up from 10% in 2021 ” Market Guide for Graph Database Management Systems, August 2022 Gartner
- 18. © 2022 Neo4j,Inc. All rights reserved. © 2022 Neo4j, Inc. All rights reserved. This seems very interesting, but how do real businesses get started? 18
- 19. © 2022 Neo4j,Inc. All rights reserved. Leading Customer Use Cases for Graph Data Science 19 Recommendations • Identified journey archetypes and patterns • Revealed journey similarities over time • Found influential touch-points across journeys Fraud Detection • 300%+ increase in fraud detection • 10% True positive alert escalations (industry is <1%) • ~150% increase in payment flow • Reduced overall number of alert escalations Customer 360 • 1621% increase touchpoint length • 500% increase visits per profile • 20-30% improvement of customer understanding Logistics / SC • Subsecond maritime routes planning • Reduce global carbon emissions 60,000 tons • 12-16M ROI for OrbitMI customers
- 20. © 2022 Neo4j,Inc. All rights reserved. The CIO’s path 20 Educate yourself Graphs are here, you need to be up to date Prepare your organization Build amazing systems Your IT staff are already tinkering with graph tech, cultivate that enthusiasm Engage with communities of practice, and vendors to map out your graph play
- 21. © 2022 Neo4j,Inc. All rights reserved. © 2022 Neo4j, Inc. All rights reserved. 21 Thanks for listening and… Dr Jesús Barrasa [email protected] Dr Jim Webber [email protected]
- 22. © 2022 Neo4j,Inc. All rights reserved. © 2022 Neo4j, Inc. All rights reserved. 22 … let’s continue the conversation at the Neo4j booth 915 D&A Marketplace Dr Jesús Barrasa [email protected] Dr Jim Webber [email protected]