Advanced Analytics in Banking, CITI
Download as PPTX, PDFâą11 likesâą7,864 views
The document discusses Citibank's strategy for adopting big data across its global operations, emphasizing a customer-centered approach and a hybrid participation model that varies by competency level. It outlines the process for developing analytics solutions, including stages from proof of concept to deployment, while also highlighting the need for a robust data governance framework. Additionally, it addresses the technical benefits of big data applications, such as enhancing algorithm efficiency and scalability for various banking analyses.
Advanced Analytics in Banking, CITI
- 1. Advanced Analytics in Banking Juan M. Huerta Global Decision Management VP Advanced Analytics Citibank
- 2. I will talk about⊠⹠Big Data Adoption process at Citi ⹠Realizing the Technical Value of Big Data ⹠Global Solutions 1
- 4. Citi: A Customer Centered Organization 3 As a customer-centered bank, the goal of our Big Data strategy to shift the focus from independent vertical silos to Common Horizontal Solutions focused around Citiâs 200-million customer accounts
- 5. Big Data Adoption Stakeholders âą Lines of Business âą Strategy & Decision Management Organizations: cross LOB & Geo, global âą Data innovation Office: Governance & Regulatory âą CitiData â Big Data & Analytics Engineering 4
- 6. Big Data Adoption Roadmap 5 Adoption will not occur at once. The level of capability maturity across the organization will vary significantly. On theory we think in terms of Staged Competencies of a Big Data Maturity Model. In practice, a hybrid process, which fits the level of maturity of participants, is needed. Common Data Common Analytic Platform Common Tools & Techniques Common Solutions Common Focus Strategy
- 7. Big Data Adoption Hybrid Participation Model âą Novice: Proof of Concept âą Expert: R&D Environment âą Shadowed 6
- 8. 7 End-to-end Analytic Process for a POC Project This is one component of the hybrid model Ideas and Hypotheses Information Asset Inventory Navigator (âIAINâ) âą Pipeline of ideas to use data for competitive advantage âą Robust, comprehensive ontology allowing analysts and economists to search, sort, and select data for analysis âą Preliminary assessment for business value, data safekeeping and alignment to business practices Data Transformation & Provisioning âą Transformation rules executed to normalize and conform production data âą Conformed data set made available in production environment Production Model Development âą Develop scalable, productizable analytics Model Deployment âą Exploit insights and analyses across the enterprise to maximize value âą Models measured for quality / usage âą Formal approval process through Business Steering Committee based on understanding expected use of production data R&D process R&D Project Approval Product Approval Engineering / Production process Analytics Knowledge Management âą Robust, compreh ensive ontology allowing analysts and economists to search, sort, an d select data for analysis Data Set Preparation & Provisioning âą Basic preparation of data set (e.g., consolidation, conformation) âą Permission-based provisioning of data set into a Big Data Analytics environment Analytics Execution âą Advanced analytic tools mine business insight from large volumes of data âą Data scientist peers review model findings and results Analytics Peer Review Data Acquisition âą Where necessary, acquire new data sets to support R&D project
- 9. Advanced Global Solutions âą A global solution is a tested algorithm or analytic model that carries out a particular business analysis and which is leveraged at a global scale âą A big data global solution enables the interplay of complex algorithms and large datasets âą When a global solution is built upon big data approaches a delivery roadmap should be considered âą In the exploratory process a Global Solution is developed in the Innovation R/D environment and validated through a POC process âą Alignment with Innovation, UAT, PRD environments 8
- 10. Technical Value of Big Data: Benchmarks and Analysis
- 11. The Boom Driving Big Data is Technological Heebyung Koh , Christopher L. Magee A functional approach for studying technological progress: Extension to energy technology Technological Forecasting and Social Change, Volume 75, Issue 6, July 2008, Pages 735â758
- 12. The Quadrant Of Analytic Opportunity Run Time is affected by Data Size and Algorithmic Complexity Algorithmic Complexity Database Interaction Mtg+Cards+ Banking Accounts Transaction features Accounts Transactions Branches Transactions Accounts Summary Stats. Employees Summary Stats. GL-GOCS GL-Entries Branches Summary Stats. 10^10 10^9 10^9 10^8 10^7 10^6 10^5 Data Size Sequence Mining Predictive filtering Latent Dirichlet Allocation HMM Baum- Welch O(ns nf nt) CART O(nf ns log ns) Iterative SVD- CF K-means Logistic Regression PCAPage Rank Self-Org. Maps Neural Nets Collaborative Filtering (CF) Vector based Approaches HMM Machine Learning Traditional Statistical Big Data/Pattern Mining Conditional Random Fields Support Vector Machines
- 13. Breaking down the gains of P13n: A Controlled Incremental Benchmark on a Workstation grade processor (x500) Implemented an incremental-SVD (Netflix Cup) predictive model that runs on midsize of datasets⊠X30 ⹠Compiled Code (vs. interpreted) x4 ⹠In Memory (vs. Disk access) X3.12 ⹠Multithread (vs. single thread) X1.3 ⹠Workstation grade processor
- 14. Basic Map Reduce Benchmarks 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 1 2 3 4 5 6 Series1 Impact of overhead as function Of input volume: Relative Map Throughput as a function of # Mappers 0 5 10 15 20 25 0 5 10 15 20 RelativeMapCPUtimespeedup Number of Maps 0.003351955 0.032258065 0.319148936 1 2.631578947 21.12676056 Linear (0.003351955 0.032258065 0.319148936 1 2.631578947 21.12676056) 0 200 400 600 800 1000 1200 1400 1600 0 5 10 15 20 TokensperWallClockSecond Number of Maps Series1 Linear (Series1)
- 15. HAMSTER: Hadoop Multi-signature Search for Text-based Entity Retrieval âą Core algorithm: String Edit Distance O(mnk2) âą Baseline runs at 100 matches per day âą HAMSTER speedup: 33x (5 node speedup) 60x (java speedup) = 2000x faster Source Items Target Items Source items per target Input Size MAP Records Cluster Max Map Tasks Effective Map Tasks CPU map (secs) Wall time 34k 618k 100 4.40GB 345 33 33 196k 2h 14 secs 34k 618k 50 8.8GB 690 40 66 196k 1h 47min 34k 618k 30 14.6GB 1,149 40 110 199k 1h 39 min
- 16. Leveraging Global Big Data Global Solutions
- 17. Creating Global Big Data solutions Our goal is to evolve from Big Data algorithms to Big Data Solutions
- 18. Example of Advanced Global Solution Matrix 17 Outlier Detection Multivariate Segmentation Sequence Matching Network Analysis Customer Contextual Clickstream Action Marketing Risk/Fraud Digital Structured Prediction 17 K-Medoids Clustering
- 19. Example: Transactional Time Series AnomalousBehavior
- 20. On Demand Simulation: Generate Branchesâ DNA âą Case Scenario: Unusual number of cash advances by 2 tellers. Single day fraud Multi day fraudOriginal branch (August)
- 21. Creating Regions of Interest based on On-Demand-Simulation Minimum-Spanning- Tree based branch association for region of interest generation Multi-day fraud simulation Original branch Region of interest âą Numbers shown are randomized indices
- 22. Conclusion: Lessons Learned âą One Size does not fit all âą Follow a Hybrid Approach âą Leverage Analytic patterns: Global Solutions âą Big Data is about Parallelization âą The future: expensive Algorithms applied to large datasets âą Global Solutions are the combination of algorithmic building blocks applied to specific business problems 21
- 23. Thank You! 22