Ant Colony Optimization (ACO) algorithm has been implemented to discover a list of classification rules. In the proposed system, a sequential covering strategy for ACO Classification algorithm used to remove the problem of interaction. In this finding the ―best‖ rule that accounts for a part of the training data. Adding the best rule to the induced rule set and removing the data it covers. This iterative process continues until no training instance remains. Heuristic function is used for selecting a best rule and calculating the predictive accuracy. In the proposed method, weather-nominal data set has been used in order to calculate the predictive accuracy.
展开▼
