关键词: 人工免疫, 投影寻踪, 降维, 优化, 聚类

Abstract: To solve the dimension reduction with Projection Pursuit (PP) model, the theory of Artificial Immune (AI) system was introduced. The immune clone-selecting algorithm was used to optimize the projecting direction, with the purpose to project the data from high dimensional space to a low one. Therefore, the projection not only reduced the computation complexity during the process of data mining, but also made the data shrinking possible. Besides, the clustering results between the initial and the processed data with K-Means and other clustering algorithms were compared. And the experimental results verify the validity of the proposed algorithm.

Key words: Artificial Immunity (AI), Projection Pursuit (PP), optimization, clustering