(Data mining) Large database for player-created teams?

Hello friends,

I'm playing with the idea of creating an unsupervised model to classify player-created Pokemon teams and sets. The idea is to use clustering analysis to group teams and/or sets into clusters which would (hopefully) represent different teambuilding archetypes (e.g. sweeper/wall/etc. or hyperoffense/stall/etc.).

This will require a substantial data set of player-created teams (maybe 300+ teams). Of course there are thousands (millions? billions??) of teams at various sources like Pokepaste and the forums here, but I need large amounts of data in a standardized format such as JSON or CSV.

Anyone know of a good resource to get this kind of data on player-created teams? I'd rather not spend hours writing a webscraper for this...

Best,
zanalytic

***Edit: the format doesn't matter as much (PS import/export format is fine), as long as all the data comply with it.
 
Last edited:
I built a parser from replay data in R and ran a team clustering algorithm on it. Most of the info is in my blog:
Clustering: https://pokemon-data-analysis.netlify.app/posts/2022-03-28-meta-cores/
Parsing (click "hide" buttons to reveal code): https://pokemon-data-analysis.netli...ically-analyzing-a-teambuild-with-regression/

You can just grind on ladder and always save your replays, put them in a single directory, and then run the parsing script. If you play alot, you may get the data you need quickly.
 

Users Who Are Viewing This Thread (Users: 1, Guests: 0)

Top