🐻♥️ for Valentine’s Day, I’m matchmaking single Cornellians!! 131 Cornellians have signed…

🐻♥️ for Valentine’s Day, I’m matchmaking single Cornellians!! 131 Cornellians have signed up so far!! we need more Cornellian Men to sign up, and more Cornellians who are flexible on Location preference!! please fill out the form in the comments below, and I’ll try to matchmake!! I’ll send an email to both Cornellians, with a short description of why the two were matched!! I asked AI to create an algorithm to help match everyone, and it’s doing a fantastic job!! "optimal total compatibility"!! ● Given your data (131 people, preference filters + text-based interests), here’s what I’d recommend: Compatibility scoring + max weight matching 1. Filter — only pair people where gender preferences are mutual (A wants B’s gender AND B wants A’s gender) 2. Score each valid pair on: - Age within preferred range (hard filter or bonus) - Location match (fuzzy string match on city/region) - Hobby overlap (Jaccard similarity on the comma-separated lists) - Activity/event overlap (same idea) 3. Match — use networkx.max_weight_matching() which finds optimal pairings on a general graph (handles same-gender preferences, not just bipartite) This is the right fit because: - 131 people is small — pairwise scoring is fine (~8500 pairs worst case) - Not strictly bipartite — some people may prefer same gender, so Gale-Shapley / Hungarian don’t cleanly apply - networkx handles general graphs with the Blossom algorithm, guaranteeing optimal total compatibility - Everyone gets at least one match — anyone unmatched after the first pass can get assigned their best remaining option