Classifying customer correctly and effectively according to customers= characteristics and behaviors plays a key role for network enterprises to realize the full values of modern customer relationship management. Aiming at the shortages of the existing particle swarm and K-means algorithm, this paper improves two algorithms through integrating them together and presents a new customer classification algorithm. First 21 customer classification indicators is designed based on consumer characteristics and behaviors analysis; Second, the study analyzes the shortages of particle swarm optimization algorithm, improves its optimal performance, redesigns a new particle swarm optimization algorithm for classifying online trade customer; Finally the experimental results verify that the new algorithm can improve effectiveness and validity of customer classification when used for classifying network trading customers practically.
展开▼
