this article, we propose to deal with this sparsity problem by applying an associative retrieval framework and related spreading activation algorithms to explore transitive associations among consumers through their past transactions and feedback. Such transitive associations are a valuable source of information to help infer consumer interests and can be explored to deal with the sparsity problem. To evaluate the effectiveness of our approach, we have conducted an experimental study using a data set from an online bookstore. We experimented with three spreading activation algorithms including a constrained Leaky Capacitor algorithm, a branch-and-bound serial symbolic search algorithm, and a Hopfield net parallel relaxation search algorithm. These algorithms were compared with several collaborative filtering approaches that do not consider the transitive associations: a simple graph search approach, two variations of the user-based approach, and an item-based approach. Our experimental results indicate that spreading activation-based approaches significantly outperformed the other collaborative filtering methods as measured by recommendation precision, recall, the F-measure, and the rank score. We also observed the over-activation effect of the spreading activation approach, that is, incorporating transitive associations with past transactional data that is not sparse may "dilute" the data used to infer user preferences and lead to degradation in recommendation performance

Abstract. Recommender systems attempt to reduce information overload and retain customers by selecting a subset of items from a universal set based on user preferences. While research in recommender systems grew out of information retrieval and filtering, the topic has steadily advanced into a legitimate and challenging research area of its own. Recommender systems have traditionally been studied from a content-based filtering vs. collaborative design perspective. Recommendations, however, are not delivered within a vacuum, but rather cast within an informal community of users and social context. Therefore, ultimately all recommender systems make connections among people and thus should be surveyed from such a perspective. This viewpoint is under-emphasized in the recommender systems literature. We therefore take a connection-oriented perspective toward recommender systems research. We posit that recommendation has an inherently social element and is ultimately intended to connect people either directly as a result of explicit user modeling or indirectly through the discovery of relationships implicit in extant data. Thus, recommender systems are characterized by how they model users to bring people together: explicitly or implicitly. Finally, user modeling and the connection-centric viewpoint raise broadening and social issues—such as evaluation, targeting, and privacy and trust—which we also briefly address. Keywords: recommendation, recommender systems, small-worlds, social networks, user modeling “What information consumes is rather obvious: it consumes the attention of its recipients. Hence a wealth of information creates a poverty of attention, and a need to allocate that attention efficiently among the overabundance of information sources that might consume it.”

Abstract. In this chapter we provide a comprehensive overview of the topic of Intelligent Techniques for Web Personalization. Web Personalization is viewed as an application of data mining and machine learning techniques to build models of user behaviour that can be applied to the task of predicting user needs and adapting future interactions with the ultimate goal of improved user satisfaction. This chapter survey’s the state-of-the-art in Web personalization. We start by providing a description of the personalization process and a classification of the current approaches to Web personalization. We discuss the various sources of data available to personalization systems, the modelling approaches employed and the current approaches to evaluating these systems. A number of challenges faced by researchers developing these systems are described as are solutions to these challenges proposed in literature. The chapter concludes with a discussion on the open challenges that must be addressed by the research community if this technology is to make a positive impact on user satisfaction with the Web. 1

Social trust relationships between users in social networks speak to the similarity in opinions between the users, both in general and in important nuanced ways. They have been used in the past to make recommendations on the web. New trust metrics allow us to easily cluster users based on trust. In this paper, we investigate the use of trust clusters as a new way of improving recommendations. Previous work on the use of clusters has shown the technique to be relatively unsuccessful, but those clusters were based on similarity rather than trust. Our results show that when trust clusters are integrated into memory-based collaborative filtering algorithms, they lead to statistically significant improvements in accuracy. In this paper we discuss our methods, experiments, results, and potential future applications of the technique.

In the research to date, the performance of recommender systems has been extensively evaluated across various dimensions. Increasingly, the issue of robustness against malicious attack is receiving attention from the research community. In previous work, we have shown that knowledge of certain domain statistics is sufficient to allow successful attacks to be mounted against recommender systems. In this paper, we examine the extent of domain knowledge that is actually required and find that, even when little such knowledge is known, it remains possible to mount successful attacks.

Recommendation and personalization attempt to reduce information overload and retain customers. While research in both recommender systems and personalization grew mainly out of information retrieval, both areas have emerged from nascent levels to veritable and challenging research areas in their own right. Whereas no technical or sophisticated methodologies exist by which to build such systems, the field also lacks a comprehensive, yet manageable survey by which to study recommenda-tion systems and personalization facilities. In this paper, we attempt to fill that gap by presenting a thematic approach toward studying recommendation and personalization. Specifically, we present three major representative personalization themes: rec-ommendation; induction, exploration, and exploitation of social networks; and personalization of information access. We unify the presentation of the three themes which we have extracted from the rich landscape of recommender system and personal-ization research via a functional metaphor, where inputs and output to a function are identified in each theme and instantiated through a number of systems and projects visited. In addition, we examine how a number of systems implement the function through various operators and techniques. Finally, we cover several broadening aspects, such as targeting, privacy and trust,

Recent years have seen an explosive growth in the quantity of information that is available. The need for automated techniques to deal with the information overload problem is clear. The ability of users to quickly locate items that meet their own specific needs is crucial. Recommender systems have now been widely and successfully implemented, particularly in e-commerce applications, as a solution to this problem. One of the most

Abstract. Recommender systems attempt to reduce information overload and retain customers by selecting a subset of items from a universal set based on user preferences. While research in recommender systems grew out of information retrieval and filtering, the topic has steadily advanced into a legitimate and challenging research area of its own. Recommender systems have traditionally been studied from a content-based filtering vs. collaborative design perspective. Recommendations, however, are not delivered within a vacuum, but rather cast within an informal community of users and social context. Therefore, ultimately all recommender systems make connections among people and thus should be surveyed from such a perspective. This viewpoint is underemphasized in the recommender systems literature. We therefore take a connection-oriented viewpoint toward recommender systems research. We posit that recommendation has an inherently social element and is ultimately intended to connect people either directly as a result of explicit user modeling or indirectly through the discovery of relationships implicit in extant data. Thus, recommender systems are characterized by how they model users to bring people together: explicitly or implicitly. Finally, user modeling and the connection-centric viewpoint raise broadening and social issues—such as evaluation, targeting, and privacy and trust—which we also briefly address.

Abstract. Recommender systems produce social networks as a side effect of predicting what users will like. However, the potential for these social networks to aid in recommending items is largely ignored. We propose a recommender system that works directly with these networks to distribute and recommend items: the informal exchange of information (word of mouth communication) is supported rather than replaced. The paper describes the push-poll approach and evaluates its performance at predicting user ratings for movies against a collaborative filtering algorithm. Overall, the push-poll approach performs significantly better while being computationally efficient and suitable for dynamic domains (e.g. recommending items from RSS feeds). 1

This paper demonstrates that "social network collaborative filtering " (SNCF), wherein user-selected like-minded alters are used to make predictions, can rival traditional user-to-user collaborative filtering (CF) in predictive accuracy. Using a unique data set from an online community where users rated items and also created social networking links specifically intended to represent likeminded “allies, ” we use SNCF and traditional CF to predict ratings by networked users. We find that SNCF using generic "friend " alters is moderately worse than the better CF techniques, but outperforms benchmarks such as byitem or by-user average rating; generic friends often are not like-minded. However, SNCF using "ally " alters is competitive with CF. These results are significant because SNCF is tremendously more computationally efficient than traditional user-user CF and may be implemented in large-scale web commerce and social networking communities. It is notoriously difficult to distinguish the contributions of social influence (where allies influence users) and "social” selection (where users are simply effective at selecting like-minded people as their allies). Nonetheless, comparing similarity over time, we do show no evidence of strong social influence among allies or friends. Categories and Subject Descriptors: