Finding Social Networks Among Online Reviewers for Customer Segmentation

Abstract

Recently, online user-generated data have emerged as a valuable source for industrial applications. In the consumer product area, many studies analyze online data and draw implications on product design. However, most of them treat online customers as one group with the same preferences, while customer segmentation is a key strategy in conventional market analysis. This paper proposes a new methodology based on text mining and network analysis for online customer segmentation. First, the method extracts customer attributes from online review data. Then, a customer network is constructed based on these attributes and predefined networking rules. For networking, a new concept of "topic similarity" is proposed to reflect social meaning in the customer network. Finally, the network is partitioned by modularity clustering, and the resultant clusters are analyzed to understand segment properties. We validate our methodology using real-world data sets of smartphone reviews. The result shows that the proposed methodology properly reflects the heterogeneity of the online customers in the segmentation result. The practical application of customer segmentation is presented, illustrating how it can help companies design target-customer-oriented products.