Data Mining And Its Applications In Mobile Business Environments

With the rapid development of mobile Internet related technologies, the number of mobile applications and services has grown at a breathtaking rate over the past few years. For example, various kinds of mobile applications and services, such as Life Entertainment, Social Network Services and Navigation, have been developed for sat-isfying different user needs. At the same time, such mobile applications and services also generate a large number of historical records of user interactions and business trans-actions, which open a venue for researchers to explore the hidden values under mobile business environment. Indeed, these business data also provide a wide range of opportu-nities and challenges for enabling novel mobile business intelligence, and thus increas-ingly attract researchers’ attention in both academia and industry. Based on the above, in this dissertation, we introduce several recent research efforts on mobile business intel-ligence based on data mining technologies. Specifically, by exploring the new business data from mobile Apps, we first propose three research directions for mobile business intelligence, namely User Understanding, Application Understanding and Application Enabling. Furthermore, from each perspective of the above directions, we propose an approach for mining the personal context-aware preferences of mobile users, an ap-proach for mobile App classification by exploiting enriched contextual information, an approach for discovering the ranking fraud of mobile Apps, an approach for modeling the popularity of mobile Apps, and an approach for mobile App recommendation with security and privacy awareness, respectively. We believe these works are significant for the healthy development and future success of mobile business intelligence. To be specific, the research contributions of this dissertation can be summarized as follows.First, we propose an approach for mining personal context-aware preferences from the context-rich device logs, or context logs for short, and exploit these identified pref-erences for building personalized context-aware recommender systems. A critical chal- lenge along this line is that the context log of each individual user may not contain sufficient data for mining his/her context-aware preferences. Therefore, we propose to first learn common context-aware preferences from the context logs of many users. Then, the preference of each user can be represented as a distribution of these com-mon context-aware preferences. Specifically, we develop two approaches for mining common context-aware preferences based on two different assumptions, namely, con-text independent and context dependent assumptions, which can fit into different ap-plication scenarios. Finally, extensive experiments on a real-world data set show that both approaches are effective and outperform baselines with respect to mining personal context-aware preferences for mobile users.Second, we propose an approach for mobile App classification with enriched con-textual information. To be specific, we first propose an approach for enriching the con-textual information of mobile Apps by exploiting the additional Web knowledge from the Web search engine. Then, inspired by the observation that different types of mobile Apps may be relevant to different real-world contexts, we also extract some contextual features for mobile Apps from the context-rich device logs of mobile users. Finally, we combine all the enriched contextual information into the Maximum Entropy model for training a mobile App classifier. To validate the proposed method, we conduct exten-sive experiments on a real-world data set to show both the effectiveness and efficiency of the proposed approach. The experimental results clearly show that our approach outperforms two state-of-the-art benchmark methods with a significant margin.Third, we provide a holistic view of ranking fraud and propose a ranking fraud de-tection system for mobile Apps. Specifically, we first propose to accurately locate the ranking fraud by mining the active periods, namely leading sessions, of mobile Apps. Such leading sessions can be leveraged for detecting the local anomaly instead of glob-al anomaly of App rankings. Furthermore, we investigate three types of evidences, i.e., ranking based evidences, rating based evidences and review based evidences, by mod-eling Apps’ ranking, rating and review behaviors through statistical hypotheses tests. In addition, we propose an optimization based aggregation method to integrate all the evidences for fraud detection. Finally, we evaluate the proposed system with real-world App data collected from the iOS App Store for a long time period. In the experiments, we validate the effectiveness of the proposed system, and show the scalability of the detection algorithm as well as some regularity of ranking fraud activities.Fourth, we propose a sequential approach based on Hidden Markov Model (H-MM) for modeling the popularity information of mobile Apps towards mobile App services. Specifically, we first propose a Popularity based HMM (PHMM) to model the sequences of the heterogeneous popularity observations of mobile Apps. Then, we introduce a bipartite based method to pre-cluster the popularity observations. This can help to learn the parameters and initial values of the PHMM efficiently. Furthermore, we demonstrate that the PHMM is a general model and can be applicable for various mobile App services, such as trend based App recommendation, rating and review spam detection, and ranking fraud detection. Finally, we validate our approach on two real-world data sets collected from the Apple Appstore. Experimental results clearly validate both the effectiveness and efficiency of the proposed popularity modeling approach.At last, we propose to develop a mobile App recommender system with priva-cy and security awareness. The design goal is to equip the recommender system with the functionality which allows to automatically detect and evaluate the security risk of mobile Apps. Then, the recommender system can provide App recommendations by considering both the Apps’popularity and the users’ security preferences. Specifically, a mobile App can lead to security risk because insecure data access permissions have been implemented in this App. Therefore, we first develop the techniques to automati-cally detect the potential security risk for each mobile App by exploiting the requested permissions. Then, we propose a flexible approach based on modern portfolio theory for recommending Apps by striking a balance between the Apps’popularity and the users’security concerns, and build an App hash tree to efficiently recommend Apps. Finally, we evaluate our approach with extensive experiments on a large-scale data set collected from Google Play. The experimental results clearly validate the effectiveness of our approach.