Association Rule Based-on Niche Genetic Algorithm for Benchmarking Management of Electric Power Company

Benchmarking management always plays an important role in electric power company, but the application of benchmarking rarely involves data mining. For resolving this issue, an association rule mining model for benchmarking management using niche genetic algorithm is developed. Based on the establishment of benchmarking indicator system, the data are gathered form Jilin Electric Power Company. With niche genetic algorithm, the association rule is optimized to maximize its support degree. Experimental results show that the performance of our approach is more promising than GA and the extracted association rules enable electric power company to find shortfalls and provide reference for the leader decision-making.