The analysis of consumption behavior pattern cluster that reflects both on-offline by region

Abstract

It is necessary to consider regional differences in consumption analysis as the consumption is highly influenced by local characteristics. Thus, this paper proposes a cluster analysis considering the local characteristics by collecting data about consumer behavior that occurs in the online and offline consumer behavior patterns. By collecting the SNS data, the consumption environment of each region is grasped, and the card payment data is collected to extract the consumption behavior pattern of the user through the sequential pattern mining. Based on this, we compute each consumption value in online and offline, and combine them. In this case, the influences from online are different according to the number of users’ SNS access. Finally, the analysis proceeds by going to a weight placed K-means clustering according to calculate the calculated value to the user area consumption. Through the proposed method in this paper, it was confirmed that the proposed method can be applied to future recommendation services.