Daniel Lemire's blog

, 1 min read

Now, that´s fun stuff!

Shane Butler is building a learner to predict AFL game results. I don’t know why, but this sounds like fun stuff!

Its been a while since my I last wrote that I was thinking of building a learner to predict AFL game results. Obviously with the finals starting this week I’m a bit late in the season but oh well! I can still simulate the season using the data I’ve collected. I’ve developed a number of Perl screen scrapers to collect all the data nessasary, which so far includes things like the match fixtures, game results, player stats and people’s tips (yes, this will be a hyprid of both historical data and human input). I’m at the stage now of using SQL generate things like a Home Ground Advantage (HGA) parameter – I guess you might call this the ETL stage!!

If he fails with his learner, he could turn this project into an OLAP project! That’d be cool!