Most existing collaborative filtering techniques have focused on modeling the binary relation of users to items by extracting from user ratings. Aside from users' ratings, their affiliated reviews often provide the rationale for their ratings and identify what aspects of the item they cared most about. We explore the rich evidence source of aspects in user reviews to improve top-N recommendation. By extracting aspects (i.e., the specific properties of items) from textual reviews, we enrich the user--item binary relation to a user--item--aspect ternary relation. We model the ternary relation as a heterogeneous tripartite graph, casting the recommendation task as one of vertex ranking. We devise a generic algorithm for ranking on tripartite graphs -- TriRank -- and specialize it for personalized recommendation. Experiments on two public review datasets show that it consistently outperforms state-of-the-art methods. Most importantly, TriRank endows the recommender system with a higher degree of explainability and transparency by modeling aspects in reviews. It allows users to interact with the system through their aspect preferences, assisting users in making informed decisions.

Commercial advertisement dissemination is one of the most promising applications in self-organizing mobile social networks (SMSN). Lightweight incentives are essential to encourage the participation of mobile users, given that only a few users are voluntary due to limited resources in mobiles. However, existing incentives overlook dishonest behaviors of relays in overstating their costs for higher payments, which in turn can reduce the revenues and utilities of disseminators. To this end, we design a sealed-auction-based incentive named DIBS. Effective algorithms utilizing lightweight information in DIBS encourage users to provide truthful information and to spread ads in differential places, where competition is less as fewer users carry the same ads in the same areas. Moreover, we prove that DIBS possesses attractive properties theoretically for auction-based incentives, i.e., lightweight, truthfulness and individual rationality. Extensive simulation results on both real and synthetic traces verify that DIBS can improve revenues and utilities of disseminators, and achieve efficient ad dissemination.

Cooperative content offloading is a promising technology to lessen heavy burden of wireless networks and improve the quality of downloading services. Since few users are voluntary in providing free assistance, auction-based incentive mechanisms are designed to encourage participation. In existing auction-based incentive mechanisms, each provider only acts as a service seller. However, a provider could also be a partner of the requestor if having interest in the requested content. This dual identity of the provider can improve the quality of its service and cut down the payment of requestor. Based on this observation, we propose an auction-based incentive mechanism named CADRE. To the best of our knowledge, CADRE is the first auction-based incentive mechanism that considers the provider's dual identity in cooperative content offloading applications. We prove that CADRE possesses attractive characteristics, i.e., truthfulness, lightweight and privacy protection. Besides, we also demonstrate that CADRE outperforms the traditional multi-attribute second-score sealed reverse auction. Our simulation results verify the theoretical analysis.

Clustering Web 2.0 items (i.e., web resources like videos, images) into semantic groups benefits many applications, such as organizing items, generating meaningful tags and improving web search. In this paper, we systematically investigate how user-generated comments can be used to improve the clustering of Web 2.0 items. In our preliminary study of Last.fm, we find that the two data sources extracted from user comments -- the textual comments and the commenting users -- provide complementary evidence to the items' intrinsic features. These sources have varying levels of quality, but we importantly we find that incorporating all three sources improves clustering. To accommodate such quality imbalance, we invoke multi-view clustering, in which each data source represents a view, aiming to best leverage the utility of different views.