How do I choose ML model?

1 pointxer2 comments

I have data and a have a problem. What are the best practices when deciding on what type of model to use for a specific problem? Flowcharts? Any no-brainers? Trial and error?

View on HN

Microsoft provides a few hints how to choose a ML algorithm depending on the prediction problem (regressio, classification, etc).

http://download.microsoft.com/download/A/6/1/A613E11E-8F9C-4...

Very nice cheat sheet!

Reply on news.ycombinator.com