Mobile video user revisit analysis based on multi-day visiting patterns

The mobile Internet is characterized by ¿Easy-come and easy-go¿ characteristics, which causes challenges for many content providers. Enclosing end users and increasing mind share of each service are crucial for service adoption. The 24-hour clickstream provides a rich opportunity to understand user's behaviors. It also raises the challenge of coping with a large amount of mobile web log data. The author examines a multi-day algorithm for user monthly-scale revisiting behavior classification for mobile video users. This was used in legacy text-oriented service in the past, however, the coverage of mobile video service users is still to be covered. In the case study section, the author shows the case studies in commercial mobile web sites and presents that the recall rate of the following month revisit prediction is approximately 80 %. The restriction of stream mining gives a small gap to the recall rates in literature, but the method has the advantage of small working memory to perform the given task of identifying the high revisit ratio users.