Using Big Data for Profiling Heavy Users in Top Video Apps
Chieh-Hsin Liao, Yu-Heng Lei, Kai-Yu Liou, Jian-Shing Lin, Hsiao-Feng Yeh · 2015
From an Internet service provider's prospective, the increasing popularity of mobile devices and broadband Internet has created new business challenges: more diverse competitions and more volatile customer behaviors. Therefore, to accurately respond to the changing customer demands, using big data to analyze existing and potential customers has become a trend among businesses in designing marketing plans and products. In this research, heavy users of 12 top video apps are identified using the network connection records in Chunghwa Telecom, Taiwan. Together with hundreds of previously extracted user features, chi-squared and ANOVA tests are performed to find features that have statistically significant differences between heavy and non-heavy users. Such profiling results can be used to design corresponding marketing plans.