Browsing by Subject "Recommender systems"
Now showing 1 - 4 of 4
Results Per Page
Sort Options
Item Creating More Credible and Likable Travel Recommender Systems: The Influence of Virtual Agents on Travel Recommender System Evaluation(2011-08-08) Yoo, Kyung HyanTo help online trip planners, some online travel agencies and travel service providers have adopted travel recommender systems. Although these systems are expected to support travelers in complex decision-making processes, they are not used efficiently by travelers due to a lack of confidence in the recommendations they provide. It is important to examine factors that can influence the likelihood of recommendations to be accepted and integrated into decision-making processes. The persuasion literature suggests that people are more likely to accept recommendations from credible and likable sources. It has also been found that technologies can be more credible and likable when they give a variety of social cues that elicit social responses from their human users. Thus, it is argued that enhancing the social aspects of travel recommender systems is important to create more persuasive systems. One approach to enhancing the social presence of recommender systems is to use a virtual agent. Current travel recommender systems use various types of virtual agents. However, it is still not clear how those virtual agents are perceived by travel recommender system users and influence users' system evaluations and interactions with these systems. Consequently, this dissertation aimed to investigate the influence of virtual agents presented in travel recommender systems on system users' perceptions. Specifically, the virtual agents' anthropomorphism as well as similarity and authority cues on system users' perceptions of system credibility and liking were examined. For this purpose, two experiments were conducted. For Study 1, the impacts of anthropomorphism of the virtual agents on users' perceptions of virtual agents as well as recommender systems in terms of credibility and attractiveness/liking were examined. Anthropomorphism was manipulated with visual human appearance and voice output. Study 2 tested the influence of virtual agents? similarity and authority on travel recommender system users' perceptions of virtual agents and system credibility and attractiveness/liking. Similarity and authority of the virtual agent were tested by manipulating nonverbal cues (age and outfit) of the agent. The results showed that the characteristics of virtual agents have some influences on system users' perceptions of virtual agents as well as recommender systems. Specifically, a human-like appearance of the virtual agent is found to positively influence users' perceived attractiveness of the virtual agent while voice outputs were found to enhance users' liking of the system (Study 1). Findings also indicate that RS users' perceptions of virtual agent expertise are increased when virtual agents wear a uniform rather than a casual outfit (Study 2). In addition, system users' perceptions of the virtual agent's credibility are found to have a significant influence on users' perceived credibility and liking of the overall system, which implies an important role of virtual agents in recommender system evaluations. Further, perceived credibility and liking of recommender systems lead to favorable evaluations of the recommendations, which, in turn, increase users' intentions to travel to the recommended destination. Past travel recommender system studies have largely neglected the social role of recommender systems as advice givers. Also, it is not clear whether the specific characteristics of virtual agents presented as a part of the system interface influence system users' perceptions. This dissertation sought to close this knowledge gap. By applying classic interpersonal communication theories to human and system relationships, this dissertation expands the scope of traditional theories used in the context of studying recommender systems. Further, the results of the research presented in this dissertation provide insights for tourism marketing as well as practical implications for travel recommender system design.Item Expertise modeling and recommendation in online question and answer forums(2009-12) Budalakoti, Suratna; Barber, K. Suzanne; ARAPOSTATHIS, ARISTOTLEQuestion and answer (Q&A) forums, as a way for seeking expertise on the Internet, have seen rapid growth in popularity in recent years. The expertise available on most such forums is voluntary, provided by individuals willing to invest their resources for no monetary remuneration. While these forums provide easy access to expertise, the expertise available is often lacking in quality and depth. Two major reasons for this are, the time investment required to participate in such forums, and the lack of a mechanism for identifying experts for specialized questions. We believe a Q&A recommender engine can ameliorate this problem significantly. The two primary contributions of this work are: a) a hierarchical Bayesian model based Q&A recommender, and b) a discussion of metrics to measure the performance of such a Q&A recommender. Two new metrics, responder load and questioner satisfaction, are suggested based on this discussion. These metrics are used to evaluate the performance of the recommender system on datasets harvested from the Yahoo! Answers website.Item Quantifying the multi-user account problem for collaborative filtering based recommender systems(2009-12) Edwards, James Adrian; Ghosh, Joydeep; Ghosh, Joydeep; Dhillon, InderjitIdentification based recommender systems make no distinction between users and accounts; all the data collected during account sessions are attributed to a single user. In reality this is not necessarily true for all accounts; several different users who have distinct, and possibly very different, preferences may access the same account. Such accounts are identified as multi-user accounts. Strangely, no serious study considering the existence of multi-user accounts in recommender systems has been undertaken. This report quantifies the affect multi-user accounts have on the predictive capabilities of recommender system, focusing on two popular collaborative filtering algorithms, the kNN user-based and item-based models. The results indicate that while the item-based model is largely resistant to multi-user account corruption the quality of predictions generated by the user-based model is significantly degraded.Item Using social network information in recommender systems(2011-08) Sudan, Nikita Maple; Ghosh, Joydeep; Baldridge, JasonRecommender Systems are used to select online information relevant to a given user. Traditional (memory based) recommenders explore the user-item rating matrix and make recommendations based on users who have rated similarly or items that have been rated similarly. With the growing popularity of social networks, recommender systems can benefit from combining history of user preferences with information from the social/trust network of users. This thesis explores two techniques of combining user-item rating history with trust network information to make better user-item rating predictions. The first approach (SCOAL [5]) simultaneously co-clusters and learns separate models for each co-cluster. The co-clustering is based on the user features as well as the rating history. This captures the intuition that certain groups of users have similar preferences for certain groups of items. The grouping of certain users is affected by the similarity in the rating behavior and the trust network. The second graph-based label propagation approach (MAD [27]) works in a transductive setting and propagates ratings of user-item pairs directly on the user social graph. We evaluate both approaches on two large public data-sets from Epinions.com and Flixster.com. The thesis is amongst the first to explore the role of distrust in rating prediction. Since distrust is not as transitive as trust i.e. an enemy's enemy need not be an enemy or a friend, distrust can't directly replace trust in trust propagation approaches. By using a low dimensional representation of the original trust network in SCOAL, we use distrust as it is and don't propagate it. Using SCOAL, we can pin-point the groups of users and the groups of items that have the same preference model. Both SCOAL and MAD are able to seamlessly integrate side information such as item-subject and item-author information into the trust based rating prediction model.