3) ChiMerge [Ker92] is a supervised, bottom-up (i.e., merge-based) data discretization method. It relies on _2 analysis: Adjacent intervals with the least _2 values are merged together until the...

Extracted text: 3) ChiMerge [Ker92] is a supervised, bottom-up (i.e., merge-based) data discretization method. It relies on _2 analysis: Adjacent intervals with the least _2 values are merged together until the chosen stopping criterion satisfies. (a) Briefly describe how ChiMerge works. (b) Take the IRIS data set, obtained from the University of California-Irvine Machine Learning Repository set to be Data (www.ics.uci.edu/_mlearn/MLRepository.html), as discretized. Perform data discretization for each of the four numeric a data attributes using the ChiMerge method. (Let the stopping criteria be: max- interval D 6). You need to write a small program to do this to avoid clumsy numerical computation. Submit your simple analysis and your test results: split-points, final intervals, and the documented source program.