Workshop - Machine Learning For Fun and Profit
summary
Claim: Ruby answers concrete user questions. Code uses sex_machine gem to infer gender (about 80% accuracy on first names), a local service built from freegeoip + MaxMind + Go to geolocate IPv4 logins, and scripts that populate a SQLite ML DB (assign_gender, assign_location). Unsupervised work uses k-means clustering and hierarchical methods via ai4r (example: clusters with means ≈12, 51, 150 badges, sizes ~61/30/others) and recommendations use SVD + cosine similarity. Everything runs but linear algebra libs sometimes fail to compile on flaky conference Wi‑Fi.
AI-written summary · about
chapters (6)
- 0:00Start
- 14:20Gender Assignment
- 27:37Location Awareness
- 39:54Tribal Culture
- 51:17Interlude: On Mathematics
- 53:39Alternatives to K-Means Hierarchical clusters
related talks
46mPractical Machine Learning and Rails
36mMachine Learning Microservices in Rails
23mFrom Data to Recommendations: Building an Intelligent System with ROR
07mUnconference pitches
01h 20mLightning Talks
26mRocky Mountain Ruby 2023 - Lightning Talks
33mHuman Powered Rails: Automated Crowdsourcing In Your RoR App
29mDraw a Crowd