Mining High Utility Mobile Sequential Patterns in Mobile Commerce Environments

Mining user behaviors in mobile environments is an emerging and important topic in data mining fields. Previous researches have combined moving paths and purchase transactions to find mobile sequential patterns. However, these patterns cannot reflect actual profits of items in transaction databases. In this work, we explore a new problem of mining high utility mobile sequential patterns by integrating mobile data mining with utility mining. To the best of our knowledge, this is the first work that combines mobility patterns with high utility patterns to find high utility mobile sequential patterns, which are mobile sequential patterns with their utilities. Two tree-based methods are proposed for mining high utility mobile sequential patterns. A series of analyses on the performance of the two algorithms are conducted through experimental evaluations. The results show that the proposed algorithms deliver better performance than the state-of-the-art one under various conditions.