英文摘要 |
Supported by the widespread use of mobile devices to access the Internet from anywhere at any time, the smartphone applications market has grown rapidly. Compared with other products, these applications may be installed and uninstalled repeatedly over time by customers. It is difficult, therefore, to acquire real data from the market. Recently, Fuller Inc., a Japanese software development company, developed an Android smartphone application called “Mr. Mobile, the battery saver,” which enables the company to obtain data concerning installments and uninstallments of smartphone applications using a legitimate agreement between the company and customers of the application. The purpose of this paper is to analyze the characteristics of smartphone applications satisfying the sustainable popularity condition defined in Sekpon et al. (2017). The paper developsseveralsegmentation algorithms to identify smartphone applications that satisfy the sustainable popularity condition based on logistic regression (LR) and decision tree (DT), with a focus on Japanese free games. Some hidden characteristics of such smartphone applications are analyzed by revealing key factors that play a significant role in the segmentation algorithms. |