Machine Learning Microservices in Rails
summary
A Rails API-only microservice crawls and tokenizes sources into a library of roughly 10–20k articles, trains a classifier with the naive Bayes gem, stores the model as a YAML snapshot, and serves top‑3 recommendations on page load by loading that file instead of recomputing. Tokenization uses the graph rank gem, background processing runs with Sidekiq, HTTP requests use HTTPParty, and the automated system returns useful recommendations that sometimes beat manual searches.
AI-written summary · about
chapters (19)
- 0:00Introduction
- 1:46Main Objective
- 2:32wireframes
- 5:21end product
- 5:58recommended guides
- 8:07passion to learn
- 8:50Mr Bowditch
- 9:57How the system was built
- 11:41Manual approach
- 12:29Venn diagram
- 13:35Gems used
- 13:56Naive Bayes
- 16:42GraphRank
- 18:35HTTP Party
- 18:55Sidekick
- 19:44Overview
- 20:22Content Crawler
- 21:36Recommendation Builder
- 24:56Filtering
related talks
27mFive machine learning techniques that you can use in your Ruby apps today By Benjamin Curtis Machine learning is everywhere these days
55mBuilding a Recommendation Engine with Machine Learning Techniques
28mMachine Learning for Fun and Profit
46mPractical Machine Learning and Rails
35mIntelligent Ruby: Getting Started with Machine Learning
23mFrom Data to Recommendations: Building an Intelligent System with ROR
42mMachine learning in your hobby project: Obstacles, entry points and solutions
51mMachine Learning with Ruby