Machine Learning for Fun and Profit
summary
Uses Ruby on your Rails Users table to produce concrete results: assigns likely gender with the sex_machine gem, backfills location with freegeoip/MaxMind lookups via Faraday, finds customer segments with k-means clustering (and notes stability issues), and builds recommendations with collaborative filtering using SVD. Code and examples are in the repo, runs in batch, and is intended to be runnable today to turn user attributes into actionable segments and simple recommendations.
AI-written summary · about
chapters (13)
- 0:00Start
- 5:14Problem 1: Gender Assignment
- 5:36Problem 1: Code
- 6:20Interesting results...
- 7:29Problem 2: Location Awareness
- 7:53Problem 2: Code
- 10:25Clustering
- 12:49Interlude: On Mathematics
- 14:15Alternatives to K-Means
- 16:17Collaboration
- 19:51Our Goal, Redux
- 23:08Tools of the Trade
- 25:23Kinds of Machine Learning
related talks
27mFive machine learning techniques that you can use in your Ruby apps today By Benjamin Curtis Machine learning is everywhere these days
46mPractical Machine Learning and Rails
23mFrom Data to Recommendations: Building an Intelligent System with ROR
35mIntelligent Ruby: Getting Started with Machine Learning
51mMachine Learning with Ruby
34mBuilding Data Driven Products with Ruby
36mMachine Learning Microservices in Rails
55mBuilding a Recommendation Engine with Machine Learning Techniques