1 in 20? Someone has extremely high standards. That being said, this was interesting, as recently I have taken in interest in applying computer science to dati

1 in 20? Someone has extremely high standards. That being said, this was interesting, as recently I have taken in interest in applying computer science to dating. I recently wrote a program that takes a list of attributes (attractiveness, intelligence, etc), then, tournament-style, matches them against each other and lets the user pick which is more important. After enough time, it has a decent idea of the relative importance of each attribute (without necessarily pairing them all, which would require N(N-1)/2 pairings, far too many for a decent sized attribute list). Then it takes as input a list of people (in my case, a list of girls that I know). It then asks questions such as "Who is more [attribute], [person1] or [person2]?". After asking a sufficient number of questions, it can produce a weighted average with which to rank people. I then treat this list as a heuristic in my metaphorical graph-search algorithm for a girlfriend. I could post the code if anyone is interested.