Strategy research of global optimization based on genetic algorithm
.基于遗传算法的全局优化检索策略研究*

Case retrieval is the key technique in CBR,and it is also the most important step in realizing artificial mining system. To improve the efficiency and accuracy of case retrieval, most traditional researches focus on attribute reduction and filter casebase, but they do not take attribute weights and K of K-NN into account. This paper proposed a new global optimization research technique based on genetic algorithm. This technique, taking advantage of genetic algorithm, can optimize case-base, attribute weights, and K of K-NN algorithm synchronously. At last,experiments show that searching efficiency and accuracy under this technique are priorer than traditional technique in efficiency and accuracy.