Extracting Insight From A Muslim Marriage App
summary
Analyzes anonymized MuzMatch data with Python, using NetworkX after neo4j crashes, community detection via Louvain modularity, and a random forest plus tf‑idf for text features. Finds a 2:1 men-to-women ratio, filters users liking >300 to get ~10,000 nodes (from ~20,000), community sizes ~1k–4k, outliers liking up to ~14,000 (men) and ~7,000 (women). Defines a popularity metric (real likes ÷ impressions) and observes typical popularity ~85% for women versus ~10% for men. Notebook and code are on GitHub.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 3:09User interactions and data goals
- 11:18Popularity and behavior patterns
- 17:23ML insights from profile data
- 18:50Graph theory in matches
- 25:07Future research directions
- 35:55Q&A
related talks
36mHappiness inside a job: a social network analysis
26mWhat Online Dating Taught Me about Algorithms
41mDemystifying Social Network Analysis (SNA)
25mDr. Steve Liu, Chief Scientist, Tinder
01h 11mBuilding a Recommendation Engine with Neo4j and Python
29mFootball (soccer) data analysis: A pedagogic introduction
37mHow quick can we be? (data visualization techniques for engineers)
27mLet's get meta! What the data from the joind.in API can teach us about the PHP