OPTIMIZATION OF ASSOCIATION RULES FOR TUBERCULOSIS USING GENETIC ALGORITHM
T Asha, S. Natarajan, K. N. Balasubramanya Murthy · International Journal of Computing · 2014
Tuberculosis (TB) is a disease caused by bacteria called Mycobacterium Tuberculosis which usually spreads through the air and attacks low immune bodies. Human Immuno deficiency Virus (HIV) patients are more likely to be attacked by TB. It is an important health problem around the world including India. Association Rule Mining is the process of discovering interesting and unexpected rules from large sets of data. This approach results in huge quantity of rules where some are interesting and others are repetitive. It also limits the quality of rules to only two measures support and confidence. In this paper we try to optimize the rules generated by Association Rule Mining for Tuberculosis using Genetic Algorithm. Our approach is to extract only a small set of high quality Tuberculosis rules among the larger set using Genetic Algorithm. In the current approach datatypes such as discrete, continuous and categorical items have been handled. The proposed experimental result includes a small set of converged TB rules that helps doctors in their diagnosis decisions. The main motivation for using Genetic Algorithms in the discovery of high-level prediction rules is that they are robust, use adaptive search techniques that perform a global search on the solution space and cope better with attribute interaction than the greedy rule induction algorithms often used in data mining.