How to make intelligent web-apps
summary
Shows how to turn raw web data into working models using Python code on GitHub: implements k-means on 100 items with iterative center updates and stops after a few iterations, uses a simple similarity metric of 1 - intersection distance (Jaccard-like), builds a supervised classifier from labeled features and a social recommender that uses co-purchase/co-friend intersections, and reports that clustering converges quickly and recommendations run fast enough for real apps.
AI-written summary · about
chapters (12)
- 0:00Intelligent web apps intro
- 1:00Why AI for web apps
- 6:33ML in web development
- 11:02Supervised vs unsupervised
- 14:25First ML model
- 21:35Clustering algorithms
- 27:45Text classification & NLP
- 32:07Recommendation systems
- 36:19Smart web frameworks
- 39:12Q&A on machine learning algorithms
- 42:44AI BoF session
- 43:38Business intelligence libraries for developers
related talks
37mBuild a Data-Driven Web App That Everyone Can Use
01h 19mDatascience on the web
36mGenerative AI + Python: Unlocking Efficiency, Personalization, and Insight
36mBuilding a Marketing Analytics Web App with Python
48mBuilding an AI/ML powered text search system
32m"How Smart is AI? – Real projects to build real understanding"
57mSurfing Digital Darwinism with AI: Thriving in the Age of Innovation
52mAI-Powered Gamification for the Web