Big Data LDN 2018: DATA SCIENCE AT ING
- 1. Data Science at Strategy, use cases, technology Bas Geerdink | 14 November 2018 @BigData_LDN #BigDataRevolution
- 2. Technology Lead at ING Analytics Academic background: A.I. & Informatics Working in IT since 2004 At ING since 2013 @bgeerdink https://nl.linkedin.com/in/geerdink Bas Geerdink
- 3. Strategy
- 4. ING is a top financial enterprise, operating since 1881 Customers 35 Million Private, Corporate and Institutional Customers Countries 40+ In Europe, Asia, Australia, North and South America Employees 51,000 Market leaders Benelux Growth markets Commercial Banking Challengers
- 5. ING’s Think Forward strategy requires a data-driven approach 1. Earn the primary relationship 2. Develop analytics skills to understand our customers better 3. Increase the pace of innovation to serve changing customer needs 4. Think beyond traditional banking to develop new services and business models Empowering people to stay a step ahead in life and in business. Simplify & Streamline Operational Excellence Performance Culture Lending Capabilities Purpose Customer Promise Strategic Priorities Enablers Creating a differentiating customer experience Clear and Easy Anytime, Anywhere Empower Keep Getting Better
- 6. Our strategy is based on the evolution of the financial industry; banking is now fully digital, traditional products are commodity 1858 20121995
- 7. We want to offer the best possible customer journey, become a ‘love brand’ and provide an integrated financial platform Secure and reliable Relevant Personal Omnichannel Predictive Actionable
- 9. § Available in the ING app after customer consent § Predict transactions, income, and future balance based on historic transactions § Fast data flow, calculations within 200ms Kijk Vooruit: Real-time Account Forecasting
- 10. Actionable Insights: small, useful notifications that help customers in their financial planning
- 11. Yolt: the smart money app § All your bank accounts together (open banking) § Payday feature § Full of Actionable Insights § Partnerships with third parties to offer additional customer benefits § It’s free! Now available in UK (yolt.com)
- 13. There is a lot of data in the bank, in many sizes and shapes
- 14. All Analytics applications follow the same pattern Structured Static Data Insight Systems of Records Streaming Event Data Real-time Customer Actions Analytics Engine historical big data fast data
- 15. The Data Lake is the “memory” of the Bank holding all data relevant for reporting and advanced analytics Metadata & Governance Catalog Core Systems Enterprise Data Warehouse Exploratory Data Archive Reports Analytics Tooling ETL Batch & Incremental Feeds
- 16. The Streaming Data Platform is a data pipeline that allows us to build use cases with fast data in a generic way Application Flow : Events: Business Flow: Business Users System configuration: Machine Learning Environment GUI (future scope) Data Scientists Feedback loop Config CEP Engine Model Execution Engine Post Processing “filter and detect pattern” “determine relevancy” “produce notification/alert” Get Customer Profile Apply Selection Criteria Score Model Detect Pattern Get Intermed Event Filter Send Output Create Intermed Event Get Business Event Create Business Event Format Event Get Raw Events Raw Event Business Event Notification Event Intermediate Event openscoring.io
- 17. Way of Working
- 18. Data Science should always aim to solve a business problem; a mapping of the outcome to the data is required 20 Affordability Classification Data Transactions Income Monthly Income Debts Monthly Expenses
- 19. Getting from raw data to a successful classification/prediction requires a step-by-step approach; the Model Process Flow 21 Import data Clean & Combine Feature extraction Model Validate Predict
- 20. Squad ING’s one way of working: business, IT, and data science go hand-in-hand SquadSquad Chapter Guild DS DA OpsOps CJE DS DA DS CJE DS Ops DS CJE Dev DevDevDevDev Tribe
- 21. Conclusions
- 22. § The banking industry is changing rapidly due to new laws, changing regulations, and ever-rising customer expectations § ING is an “IT company with a banking licence”, whose competitive advantage is in its use of data and analytics § The data lake is the memory of the bank, providing secure and reliable access to all data § ING build a streaming data platform as a generic, multi-purpose solution with fraud detection and actionable insights as it first use cases § AI technology gives us the power to create innovative ‘beyond banking’ FinTech services, such as “Kijk Vooruit”, Yolt, and robo advisers § We work agile, follow DevOps principles, combine business and IT, and set up a data community within ING ING is a data driven enterprise, with data science at its core
- 23. Have a nice future! (we’re hiring!)
- 24. Follow us to stay a step ahead YouTube.com/ING SlideShare.net/ING@ING_News LinkedIn.com/company/ING Flickr.com/INGGroupFacebook.com/ING