This document discusses using data mining techniques like decision trees, CN2 rule, SOM, and K-means clustering to predict heart disease. It provides background on heart disease prevalence and risk factors. The methodology section describes how classification trees, CN2 rule induction, self-organizing maps (SOM), and K-means clustering algorithms work and a comparative study is performed on heart disease data to evaluate the accuracy of each technique. Experimental results show CN2 rule and SOM achieved the highest classification accuracy rates above 93%.