Cluster Analysis of Consumer's Behaviors Based on Unsupervised Learning

In recent years, the rapid development of e-commerce caused a lot of commodity information and transaction information used ineffectively. Motivated by this observation, the store number, consumption date, consumption time, consumption amount, and so on as the characteristics of consumer behavior, are taken into account. Through the analysis of these characteristics, the consumer behavior is classified to achieve the purpose of intelligent recommendation of goods according to the clustering characteristics. The experimental results demonstrate that the proposed method is effective and correct.