Class Activity 17: More decision tree implementation

Complete the evaluation() and findBestThreshold() methods. The evaluation() method should use information gain as explained and the findBestThreshold() should use the algorithm described in the GI17 videos. If you finish within 65 minutes, test your code using the SMS spam data set. You should be able to use the kNN evaluation code to evaluate the performance of your decision tree.

There is solution code for Class Activity 16 and GI17 on the class Dropbox in the gi17-solutions folder.

Work on completing this for 65 minutes. Submit the files you modified after that time.