Study of Analysis Methods for Startup Log Data of Smartphone Applications

Abstract

In this paper, using the application launch log data (10 segments of each gender age) of 5000 smartphone users (about 2.4 million lines for one week in January 2019), we conducted an analysis to identify the characteristics and differences of smartphone usage in each gender age. In this study, we conducted a market basket analysis using the Apriori algorithm in "R" to reveal the access transitions in the application, and compared the network structure of each segment in parameta of confidence and support. Results of fine tuning of the parameters, we were able to remove a very small number of applications with extremely high number of launches, or many applications with extremely low number of launches as results. In this paper, we will discuss the differences in the apps used by each age groups and gender.