改进遗传算法在文本聚类中的应用研究

RESEARCH ON THE APPLICATION OF TEXT CLUSTERING ALGORITHM BASED ON IMPROVED GENETIC ALGORITHM

分析了K均值聚类算法(K-means)存在的不足和改进遗传算法的全局优化能力,提出一种基于改进遗传算法的文本聚类方法,该方法将原始文档转化成用向量空间模型来描述的文本向量,首先随机产生若干个文档向量作为初始聚类中心形成遗传算法的染色体种群,经过改进遗传算法的选择、交叉、变异进化运算,得到较为优化的K均值聚类算法的初始聚类中心。实验表明该算法文本聚类提高了查准率和查全率,算法的高效性也得到了验证。

A text clustering method based on improved genetic algorithm is presented after analyzing the disadvantages of KMeans algorithm and the global optimization ability of improved genetic algorithm.In this method,the original document is set into text vector by a vector space model.First,we chose initial clustering centers randomly to form chromosomes populations of genetic algorithm among document vectors.The optimized K-means clustering algorithm initial cluster centers were obtained by means of selection,crossover and mutation of improved genetic algorithm.The experiments show that the text clustering algorithm improves the precision and the recall rate and the efficiency of the algorithm has been verified.