K-Means Cluster Analysis Based on Consumer Behavior

Consumer behavior analysis in this society has a very important role, use kaggle prospective analysis of open datasets, hoping to provide reference for future research. First, we can using the k-means clustering algorithm to build a model for different category of consumer, and explore associations between data points based on clustering, and give platform builders a certain reference from a commercial point of view. The experimental results show that the method of using business intelligence can well predict the future consumption tendency of different consumer groups. If the sample size were able continuously increasing, feature point and a multidimensional matrix can be formed with sample, K-means algorithm will achieve better results.