Parallel fuzzy rule learning using an ACO-based algorithm for medical data mining

Ant Colony Optimization (ACO), an inspired algorithm from nature, has been successfully applied to classification tasks of data mining in recent years. This paper proposes a rule-based system for medical data mining by using a combination of ACO and fuzzy set theory, named FACO-Miner. FACO-Miner utilizes an ACO algorithm to learn a set of fuzzy rules from labeled data in parallel manner which causes to reduce the computation time to build classifier. To detect the Don't Care (DC) attributes we have proposed a new heuristic information formula which measures the uniformity of attributes domain to find DC probability. Also, FACO-Miner has some new features that make it different from existing classifiers based on ACO meta-heuristic. To classify test samples we have defined the new fuzzy reasoning method based on averaging which takes account both the number of rules and the covering value to classify the input samples. To evaluate the performance of FACO-Miner, we use several well-known medical data sets from UCI repository. Our experiments have confirmed that FACO-Miner leads us to significant results and outperforms several famous methods in classification accuracy for medical classification.