Jeremy Howard gave a talk at the Melbourne R User Group on 16th March 2011.
Jeremy provided tips on how to successfully compete in data mining competitions. He showed how he combines R with other tools to build predictive models. He gave a walkthrough of the data, visualizations, and code, for a number of his competition entries. The talk also included an introduction to the theory behind Jeremy’s favourite modelling algorithm: random forests.
Screencast of the talk:
- Data mining competitions at Kaggle
- Official meetup description
- @jeremyphoward on Twitter
- Additional resources on Jeremy Howard’s homepage