Enhanced Ant Colony System for Water Quality Rules Classification in Smart Aquaculture

Sensed data from aquaculture monitoring will be classified and analyzed in order to determine the water quality that is suitable for the process of breeding, raising, and harvesting aquatic species. Managing huge amounts of data that are continuously sensed in real time and have different attributes in each instance of sequence requires systematic data classification which involves discovering patterns and connections in massive amounts of information. Multiple optimization problems, including data classification, have successfully been resolved with the Ant Colony System (ACS). An enhanced ACS algorithm for rule-based data classification in smart aquaculture is proposed in this study to offer a more feasible and effective solution. The benefits of rule-based data classification are combined with the ACS to optimize the accuracy and number of rules. The enhanced pheromone updating method can optimize convergence and accuracy of data classification in smart aquaculture systems. The proposed algorithm was evaluated against the other ant-based data classification algorithms and the results showed significant improvement in terms of accuracy, number of rules, and convergence time. The proposed ACS-E improved up to 85% for classification accuracy, 6.2 for the number of discovered rules and attained 10% faster compared to other ant-based classification algorithm.