Shyong K. Lam

MovieLens unplugged: experiences with an occasionally connected recommender system

Recommender systems have changed the way people shop online. Recommender systems on wireless mobile devices may have the same impact on the way people shop in stores. We present our experience with implementing a recommender system on a PDA that is occasionally connected to the network. This interface helps users of the MovieLens movie recommendation service select movies to rent, buy, or see while away from their computer. The results of a nine month field study show that although there are several challenges to overcome, mobile recommender systems have the potential to provide value to their users today

Shilling recommender systems for fun and profit

Recommender systems have emerged in the past several years as an effective way to help people cope with the problem of information overload. One application in which they have become particularly common is in e-commerce, where recommendation of items can often help a customer find what she is interested in and, therefore can help drive sales. Unscrupulous producers in the never-ending quest for market penetration may find it profitable to shill recommender systems by lying to the systems in order to have their products recommended more often than those of their competitors. This paper explores four open questions that may affect the effectiveness of such shilling attacks: which recommender algorithm is being used, whether the application is producing recommendations or predictions, how detectable the attacks are by the operator of the system, and what the properties are of the items being attacked. The questions are explored experimentally on a large data set of movie ratings. Taken together, the results of the paper suggest that new ways must be used to evaluate and detect shilling attacks on recommender systems.

Getting to know you: learning new user preferences in recommender systems

Recommender systems have become valuable resources for users seeking intelligent ways to search through the enormous volume of information available to them. One crucial unsolved problem for recommender systems is how best to learn about a new user. In this paper we study six techniques that collaborative filtering recommender systems can use to learn about new users. These techniques select a sequence of items for the collaborative filtering system to present to each new user for rating. The techniques include the use of information theory to select the items that will give the most value to the recommender system, aggregate statistics to select the items the user is most likely to have an opinion about, balanced techniques that seek to maximize the expected number of bits learned per presented item, and personalized techniques that predict which items a user will have an opinion about. We study the techniques thru offline experiments with a large pre-existing user data set, and thru a live experiment with over 300 users. We show that the choice of learning technique significantly affects the user experience, in both the user effort and the accuracy of the resulting predictions.

On the recommending of citations for research papers

Collaborative filtering has proven to be valuable for recommending items in many different domains. In this paper, we explore the use of collaborative filtering to recommend research papers, using the citation web between papers to create the ratings matrix. Specifically, we tested the ability of collaborative filtering to recommend citations that would be suitable additional references for a target research paper. We investigated six algorithms for selecting citations, evaluating them through offline experiments against a database of over 186,000 research papers contained in ResearchIndex. We also performed an online experiment with over 120 users to gauge user opinion of the effectiveness of the algorithms and of the utility of such recommendations for common research tasks. We found large differences in the accuracy of the algorithms in the offline experiment, especially when balanced for coverage. In the online experiment, users felt they received quality recommendations, and were enthusiastic about the idea of receiving recommendations in this domain.

Creating, destroying, and restoring value in wikipedia

Wikipedias brilliance and curse is that any user can edit any of the encyclopedia entries. We introduce the notion of the impact of an edit, measured by the number of times the edited version is viewed. Using several datasets, including recent logs of all article views, we show that an overwhelming majority of the viewed words were written by frequent editors and that this majority is increasing. Similarly, using the same impact measure, we show that the probability of a typical article view being damaged is small but increasing, and we present empirically grounded classes of damage. Finally, we make policy recommendations for Wikipedia and other wikis in light of these findings.

WP:clubhouse?: an exploration of Wikipedia's gender imbalance

Wikipedia has rapidly become an invaluable destination for millions of information-seeking users. However, media reports suggest an important challenge: only a small fraction of Wikipedias legion of volunteer editors are female. In the current work, we present a scientific exploration of the gender imbalance in the English Wikipedias population of editors. We look at the nature of the imbalance itself, its effects on the quality of the encyclopedia, and several conflict-related factors that may be contributing to the gender gap. Our findings confirm the presence of a large gender gap among editors and a corresponding gender-oriented disparity in the content of Wikipedias articles. Further, we find evidence hinting at a culture that may be resistant to female participation.

Interfaces for eliciting new user preferences in recommender systems

Recommender systems build user models to help users find the items they will find most interesting from among many available items. One way to build such a model is to ask the user to rate a selection of items. The choice of items selected affects the quality of the user model generated. In this paper, we explore the effects of letting the user participate in choosing the items that are used to develop the model. We compared three interfaces to elicit information from new users: having the system choose items for users to rate, asking the users to choose items themselves, and a mixed-initiative interface that combines the other two methods. We found that the two pure interfaces both produced accurate user models, but that directly asking users for items to rate increases user loyalty in the system. Ironically, this increased loyalty comes despite a lengthier signup process. The mixed-initiative interface is not a reasonable compromise as it created less accurate user models with no increase in loyalty.

MovieLens Unplugged: Experiences with a Recommender System on Four Mobile Devices

Recommender systems have changed the way people shop online. Recommender systems on wireless mobile devices may have the same impact on the way people shop in stores. There are several important challenges that interface designers must overcome on mobile devices: Providing sufficient value to attract prospective wireless users, handling occasionally connected devices, privacy and security, and surmounting the physical limitations of the devices. We present our experience with the implementation of a wireless movie recommender system on a cellphone browser, an AvantGo channel, a wireless PDA, and a voice-only phone interface. These interfaces help MovieLens users select movies to rent, buy, or see while away from their computer. The results of a nine month field study show that although wireless has still not arrived for the majority of users, mobile recommender systems have the potential to provide value to their users today.

Is Wikipedia growing a longer tail

Wikipedia has millions of articles, many of which receive little attention. One group of Wikipedians believes these obscure entries should be removed because they are uninteresting and neglected; these are the deletionists. Other Wikipedians disagree, arguing that this long tail of articles is precisely Wikipedias advantage over other encyclopedias; these are the inclusionists. This paper looks at two overarching questions on the debate between deletionists and inclusionists: (1) What are the implications to the long tail of the evolving standards for article birth and death? (2) How is viewership affected by the decreasing notability of articles in the long tail? The answers to five detailed research questions that are inspired by these overarching questions should help better frame this debate and provide insight into how Wikipedia is evolving.

The effects of group composition on decision quality in a social production community

Online social production communities allow efficient construction of valuable and high-quality information sources. To be successful, community members must be effective at collaboration, including makink collective decisions in the presence of disagreement. We examined over 100,000 decisions made by small working groups in Wikipedia, and analyzed how decision quality in these online groups is influenced by four group composition factors: the size of the group, how members were invited to the group, the prior experience of group members, and apparent bias shown by the group administrator. Our findings lead us to recommendations for designers of social production communities.