Smart Banking - Real Time Driven
summary
Number26 describes a concrete pipeline: real time categorisation of transactions using AWS Kinesis, DynamoDB and Amazon Machine Learning plus NLP (stanford nltk) and scikit‑learn classifiers, and predictive modelling of monthly finances with Keras neural nets. The system processes ~15 million transactions, deploys via Docker on Elastic Beanstalk with Jenkins CI, retrains nightly and stores predictions (not per‑user models), reports prediction intervals, and shows a working, scalable classifier and forecasting service with measured accuracy.
AI-written summary · about
chapters (15)
- 0:00Smart banking introduction
- 1:43Real-time banking necessity
- 2:47Python for personalized banking
- 3:50BOTS island instant cash
- 4:55Overdraft and financial planning
- 5:38Predicting financial status
- 7:00ML for transaction classification
- 8:03NLP for transaction texts
- 13:19Neural networks predictions
- 16:57Keras models tuning
- 18:45Future balance prediction
- 20:31AWS and Kubernetes scaling
- 23:23Docker and Elastic Beanstalk deploy
- 27:03Smart banking summary
- 28:32Q&A deployment strategies
related talks
39mReal-time Transaction Categorization w/ Bayesian Feedback Loop
13mTransforming Financial Services with Real-Time Analytics
16mNext-Gen Tech for Next-Gen Banking
35mBank evolution to data science-driven business model
36mDigital, data-driven financial advisory
16mOperationalizing AI in Finance: Real-Time Analytics
31mBuilding Data-Driven Client Relationship Management in Banking with Python
01h 03mAI Innovation for a Smarter Financial Ecosystem