A work around would be to bin the data, that is divide it into some range lets say 0-5, 5-10,10-15
and consider each range to be a feature. If the given value corresponds to a range, the value for that feature would be 1. A better solution is to use something like SVM, that can be used much more easily with continuous data. Chirag Nagpal Department of Computer Engineering Army Institute of Technology, Pune www.chiragnagpal.com ________________________________________ From: Deeksha Sharma <dsha...@scu.edu> Sent: Monday, March 2, 2015 2:09 AM To: user@mahout.apache.org Subject: Naive Bayesian Classifier on Numerical Data Hi, I am working on a data set that looks like below. The last column is the class label. -98,85,85,cold 61,80,90,hot 58,83,86,overcast 45,70,96,overcast 90,68,80,hot 10,65,70,overcast 88,64,65, hot 86,72,95, hot 22,69,70,overcast 35,75,80,overcast -45,75,70,cold 77,72,90,hot -88,81,75,cold 0,71,91,cold I am planning to use Naive Bayesian Classifier , however all my features are numeric. I am not able to find an example of how to use Naive Bayesian Classifier for numerical data. Can someone guide ?
