Applying Data Mining technology in Imperialist Competitive Algorithm for solving Traveling Salesman Problem

• Main Authors: Chen-Chuan Li, 李鎮全
• Other Authors: Pei-Chann Chang
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/ejyf6h

碩士 === 元智大學 === 資訊管理學系 === 104 === Evolutionary algorithms are often use to solve the combination problems. In recent years, researcher used heuristic algorithms to solve optimization problem. Most of the algorithms can achieve a good solution quality, but due to the combination of problems the complexity of solution is gradually increased, therefore, the evolution algorithm is likely to occur premature convergence or loss group of ethnic diversity. This study presents the DMICA (Applying Data Mining technology in Imperialist Competitive Algorithm). The concept of ICA (Imperial Competitive Algorithm) is based on random combination of ways to generate the initial solution, followed by K-means clustering algorithm. Each feasible solution is in accordance to the algorithm hamming distance, respectively with the center of each cluster. The information gain method used to identify the key priority position for mining blocks; we assemble these blocks to make an artificial chromosome. To verify the quality and stability of proposed DMICA, we applied proposed approach on benchmark problem of TSP. The results show that the proposed method is a competitive type of evolutionary algorithm.