[4C1-GS-11] AI and Society:

[4C1-GS-11-02] Predictability and linkage of returnee users in C2C market

Ryo Hatanaka¹, Ryuji Watanabe², Yusuke Miyake², 〇Fujio Toriumi¹ (1. The University of Tokyo, 2. Pepabo R&D Institute, GMO Pepabo, Inc.)

Keywords:C2C market, marketing, big data, user behavior analysis, returnee users

For service providers, users who have used the service but are currently inactive are important because they may resume using it. However, there has been few detailed research on users’ return. In this study, we focused on "returnee users" and analyzed the return of buyers and sellers in the handmade C2C market, minne. First, an experiment was conducted to predict whether buyers and sellers, who had not engaged in transactions for a long period, would purchase or exhibit products again. The results showed very high accuracy for buyers, but not for sellers Additionally, the contribution of each explanatory variable to the predictions revealed that there are some variables that significantly contribute to the prediction of return. During the experiments, the existence of interdependence between sellers’ return and buyers’ return was suggested, so we verified this. The results showed that among buyers who had not made purchases for a while, there were some people who had been waiting for their liked sellers to exhibit new products, and the return of the sellers triggered their return.