Main Article Content
Supported by the wide spread of mobile devices to access the Internet in a ubiquitous manner from anywhere at any time, the smartphone application market has been growing rapidly. In comparison with ordinary products, smartphone applications are peculiar in that they may be installed and uninstalled dynamically over time by customers in a repeated manner. Because of this, it is quite difficult to acquire real data from the market. Recently, a Japanese software development company named Fuller, Inc. developed an Android smartphone application named ‘Mr. Mobile, the battery saver,’ which enabled the company to obtain actual data concerning installments and un-installments of smartphone applications via the legitimate agreement between the company and the customers of this application. Based on the large volume of data acquired from Fuller, Inc., the purpose of this paper is to analyze the characteristics of smartphone applications satisfying the sustainable popularity condition defined in the following manner as in Sekpon et al. (2017). Let be the month in which application ‘ ’ is introduced into the market. Then application ‘a’ is said to satisfy the sustainable popularity condition associated with if it is ranked within top by Google Play at least once during the first months since , and is still ranked within top at least once during the months after . The thrust of this paper is to develop several segmentation algorithms for identifying smartphone applications satisfying the sustainable popularity condition based on Logistic Regression (LR) and Decision Tree (DT) with focus on Japanese free games. Some hidden characteristics of such smartphone applications would be analyzed by revealing key factors that play a significant role in the segmentation algorithms. The results obtained in this paper provide subtle and useful information for software development companies.