关键词: 改进粒子群, 智能优化, 粗糙集, 连续属性离散化

Abstract: An algorithm of continuous attribute discretization based on improved particle swarm is proposed. The algorithm combines intelligent optimization theory and rough sets theory. Each attribute discretization points are initialized to particle group, seeking the optimal discretization points through the interaction between particles. Discretization algorithm is applied to the UCI data sets in the experiment, and the experimental results show that, the algorithm can make the loss of information decision system reduce to the minimum, and get more concise decision rules.

Key words: improved particle swarm, intelligent optimization, rough sets, continuous attribute discretization