【作者】 王冬；

【导师】 程晓荣；

【作者基本信息】 华北电力大学（河北） ， 计算机应用技术， 2010， 硕士

【摘要】 粒子群算法是一种新型的进化计算技术。本文首先对中文分词技术和文本特征选择技术进行了深入研究,分析了这些技术的原理以及基本步骤,以及一些常见的算法,根据本文特点,对已有的一些相关算法进行了改进。接着本文对PSO算法进行了研究分析,算法的介绍、原理、基本步骤、应用步骤以及参数设置等内容。引入惯性因子和收敛因子对算法的寻优能力因素进行了实验,并对结果进行了分析。根据粒子群算法的特点以及分类规则挖掘方法,设计了粒子群分类器,并将其应用到信息过滤中,结果能够有效的对信息进行过滤,由此证明了PSO算法在处理信息过滤问题时的可行性和有效性。更多还原

【Abstract】 Particle Swarm Optimization is a new kind of evolutionary computation. In this paper we first study the Chinese word segmentation and text features selection technologies, and analyze the principles and basic steps of these technologies as well as some common algorithms. Moreover, some related algorithms are improved according to the characteristic of this paper. Then we analyze the PSO algorithm, including the introduction、principles、basic steps、application procedures and parameter setting, etc. Inertial factor and convergence factor are introduced to test the optimal capacity of the algorithm in the experiment and the experimental results are also analyzed. We design a PSO classifier according to the characteristic of PSO and the mining method of classification rules, and apply the classifier to information filtering. Results show that this proposed classifier performs well when filtering information. Thus we can prove that PSO is promising in terms of feasibility and validity of information filtering. Wang Dong(Computer Applied Technology)更多还原