摘要
针对KNN的K值难以确定的问题,提出一种基于并行遗传算法的KNN分类方法.该方法采用粗粒度模型的并行遗传算法进行设计,通过种群内的遗传、变异和种群间的并行进化、联姻得到优化的K值和分类结果.实验结果表明,该方法有效的提高了KNN算法的分类效果,是一种精确高效的分类方法.
KNN（K-Nearest Neighbour） is one of the best text classification algorithms in vector space model. The classification result of KNN depends on the choice of K value in a large extent. To the question that K value of KNN is difficult to determine, the author proposes a KNN Classification Method Based on Parallel Genetic Method. In the method the optimum K value and optimum result of classification are obtained by means of heredity, mutation in the community, and parallel evolution, intermarriage among communities. Experiments show that this method, improving the classification result of KNN availably, is an accurate and effective classification method.
出处
《西南师范大学学报（自然科学版）》
CAS
CSCD
北大核心
2010年第2期103-106,共4页
Journal of Southwest China Normal University(Natural Science Edition)
关键词
并行遗传算法
KNN算法
分类
parallel genetic algorithm
KNN algorithm
classification
