基于粒子群算法的投影寻踪分类模型研究

Research on projection pursuit classification model based on particle swarm optimization algorithm

投影寻踪分类模型将高维分析问题的数据投影到最佳投影方向上,将其转化为一维问题进行分析研究,其实质是一种降维处理技术,以达到在低维空间分析高维非线性数据的目的.寻找最佳投影方向是一个优化搜索过程,PSO算法可以有效地应用于投影寻踪分类模型中最佳投影方向的搜索.用最佳投影方向计算样本数据的最佳投影值,根据K-均值聚类算法对投影值进行聚类,获取聚类结果.

Projection Pursuit Classification Model is performed by high dimensional analysis of the data in the best projection direction, it can be translated into one-dimensional analysis of the issue, its essence is a dimensionality reduction technology in order to meet the purpose of analyzing high-dimensional and non-linear data in the low-dimensional space. The best projection direction is found to be a projection optimization search process. PSO algorithm can be effectively applied to projection pursuit classification model in order to find the best projection direction. With the best projection direction computation the cluster according to the K-means clustering algorithm to the projection value is carried out, and clustering results are obtained.