So looking at our problem, we have 10 columns of numerical data. We have no domain knowledge (or at least I haven't seen any provided by wikle) on what the columns represent. We also have a outcome variable that is categorical, with 9 different categories, that again, we have no domain knowledge as to what the values mean.
As a first stab at this problem, we should figure out what methods might be worth giving a shot?
I'm reading through the comments on project 1 that Wikle posted and hopefully those will have some good recommendations for how to approach this sort of thing.
At first blush however, I'm thinking at least we should try one tree based method for classification!
All,
So looking at our problem, we have 10 columns of numerical data. We have no domain knowledge (or at least I haven't seen any provided by wikle) on what the columns represent. We also have a outcome variable that is categorical, with 9 different categories, that again, we have no domain knowledge as to what the values mean.
As a first stab at this problem, we should figure out what methods might be worth giving a shot?
I'm reading through the comments on project 1 that Wikle posted and hopefully those will have some good recommendations for how to approach this sort of thing.
At first blush however, I'm thinking at least we should try one tree based method for classification!