Data science with angular
summary
Claims a working pipeline that senses user context and reorders page components to maximize interaction, demonstrated by a landing page where a signup box moves to the middle after repeated clicks. Front end uses Angular 2 + TypeScript with a custom smart-stack directive and smart-variant components that send click metrics; backend is Node proxying Google Analytics, streaming through Kafka + Spark, analyzed in IPython with scikit-learn using k-means and linear regression (q-learning mentioned). Models feed back to the app and the demo shows the approach successfully updating layout in real time.
AI-written summary · about
chapters (8)
- 0:00Introduction and setup
- 1:52Fetching Google Analytics data
- 8:03Why use computers for data science
- 9:49Four steps to smart components
- 13:04Demo of smart variant components
- 16:14Machine learning for user interaction
- 19:55Optimizing conversion rates
- 21:41Future of smart Angular apps
related talks
21mBuild, Measure, and Machine Learn with Angular
41mRealtime "Eye Candy" with AngularJS
14mFuture-proof web apps with Angular and TensorFlow.js
32mLeveraging Angular-Based Frameworks for AI-Driven Testing
38mStorytelling with your data
22mHow to make robust and scalable front-end using layers
38mIntroduction to AngularJS
47mAngular in the Jamstack