I have to write code to implement regression using random forests (by default Weka provides random forests for classification). Is this possible to do?
It appears that WEKA 3.7's RandomForest implementation supports a numeric class attribute, whereas 3.6 did not.
- Solved – Custom error function for randomForest R package
- Solved – How does Weka combine the decision trees in a random forest
- Solved – Should using SMO classification in WEKA take so long with large dataset
- Solved – way to remove individual trees from a forest in the randomForest package in R
- Solved – “Random Forests” or “Random Forest”