Publicly accessible adaptive systems such as collaborative recommender systems present a security problem. Attackers, who cannot be readily distinguished from ordinary users, may inject biased profiles in an attempt to force a system to “adapt ” in a manner advantageous to them. Such attacks may lead to a degradation of user trust in the objectivity and accuracy of the system. Recent research has begun to examine the vulnerabilities and robustness of different collaborative recommendation techniques in the face of “profile injection ” attacks. In this article, we outline some of the major issues in building secure recommender systems, concentrating in particular on the modeling of attacks and their impact on various recommendation algorithms. We introduce several new attack models and perform extensive simulation-based evaluations to show which attacks are most successful and practical against common recommendation techniques. Our study shows that both user-based and item-based algorithms are highly vulnerable to specific attack models, but that hybrid algorithms may provide a higher degree of robustness. Using our formal characterization of attack models, we also introduce a novel classification-based approach for detecting attack profiles and evaluate its effectiveness in neutralizing attacks.

Recommender Systems (RS) suggests to users items they will like based on their past opinions. Collaborative Filtering (CF) is the most used technique to assess user similarity between users but very often the sparseness of user profiles prevents the computation. Moreover CF doesn't take into account the reliability of the other users. In this paper we present a real world application, namely moleskiing.it, in which both of these conditions are critic to deliver personalized recommendations. A blog oriented architecture collects user experiences on ski mountaineering and their opinions on other users. Exploitation of Trust Metrics allows to present only relevant and reliable information according to the user's personal point of view of other authors trustworthiness. Di#erently from the notion of authority, we claim that trustworthiness is a user centered notion that requires the computation of personalized metrics. We also present an open information exchange architecture that makes use of Semantic Web formats to guarantee interoperability between ski mountaineering communities.

Abstract. In this chapter we present an overview of Web personalization process viewed as an application of data mining requiring support for all the phases of a typical data mining cycle. These phases include data collection and preprocessing, pattern discovery and evaluation, and finally applying the discovered knowledge in real-time to mediate between the user and the Web. This view of the personalization process provides added flexibility in leveraging multiple data sources and in effectively using the discovered models in an automatic personalization system. The chapter provides a detailed discussion of a host of activities and techniques used at different stages of this cycle, including the preprocessing and integration of data from multiple sources, as well as pattern discovery techniques that are typically applied to this data. We consider a number of classes of data mining algorithms used particularly for Web personalization, including techniques based on clustering, association rule discovery, sequential pattern mining, Markov models, and probabilistic mixture and hidden (latent) variable models. Finally, we discuss hybrid data mining frameworks that leverage data from a variety

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

As an indispensable technique in the field of Information Filtering, Recommender System has been well studied and developed both in academia and in industry recently. However, most of current recommender systems suffer the following problems: (1) The large-scale and sparse data of the user-item matrix seriously affect the recommendation quality. As a result, most of the recommender systems cannot easily deal with users who have made very few ratings. (2) The traditional recommender systems assume that all the users are independent and identically distributed; this assumption ignores the connections among users, which is not consistent with the real world recommendations. Aiming at modeling recommender systems more accurately and realistically, we propose a novel probabilistic factor analysis framework, which naturally fuses the users ’ tastes and their trusted friends ’ favors together. In this framework, we coin the term Social Trust Ensemble to represent the formulation of the social trust restrictions on the recommender systems. The complexity analysis indicates that our approach can be applied to very large datasets since it scales linearly with the number of observations, while the experimental results show that our method performs better than the state-of-theart approaches.

Data sparsity, scalability and prediction quality have been recognized as the three most crucial challenges that every collaborative filtering algorithm or recommender system confronts. Many existing approaches to recommender systems can neither handle very large datasets nor easily deal with users who have made very few ratings or even none at all. Moreover, traditional recommender systems assume that all the users are independent and identically distributed; this assumption ignores the social interactions or connections among users. In view of the exponential growth of information generated by online social networks, social network analysis is becoming important for many Web applications. Following the intuition that a person’s social network will affect personal behaviors on the Web, this paper proposes a factor analysis approach based on probabilistic matrix factorization to solve the data sparsity and poor prediction accuracy problems by employing both users ’ social network information and rating records. The complexity analysis indicates that our approach can be applied to very large datasets since it scales linearly with the number of observations, while the experimental results shows that our method performs much better than the state-of-the-art approaches, especially in the circumstance that users have made few or no ratings.

In this paper, we propose a framework that enables the detection of noise in recommender system databases. We consider two classes of noise: natural and malicious noise. The issue of natural noise arises from imperfect user behaviour (e.g. erroneous/careless preference selection) and the various rating collection processes that are employed. Malicious noise concerns the deliberate attempt to bias system output in some particular manner. We argue that both classes of noise are important and can adversely e#ect recommendation performance. Our objective is to devise techniques that enable system administrators to identify and remove from the recommendation process any such noise that is present in the data. We provide an empirical evaluation of our approach and demonstrate that it is successful with respect to key performance indicators.

Abstract. Web-based environments typically span interactions between humans and software services. The management and automatic calculation of trust are among the key challenges of the future service-oriented Web. Trust management systems in large-scale systems, for example, social networks or service-oriented environments determine trust between actors by either collecting manual feedback ratings or by mining their interactions. However, most systems do not support bootstrapping of trust. In this paper we propose techniques and algorithms enabling the prediction of trust even when only few or no ratings have been collected or interactions captured. We introduce the concepts of mirroring and teleportation of trust facilitating the evolution of cooperation between various actors. We assume a user-centric environment, where actors express their opinions, interests and expertises by selecting and tagging resources. We take this information to construct tagging profiles, whose similarities are utilized to predict potential trust relations. Most existing similarity approaches split the three-dimensional relations between users, resources, and tags, to create and compare general tagging profiles directly. Instead, our algorithms consider (i) the understandings and interests of actors in tailored subsets of resources and (ii) the similarity of resources from a certain actor-group’s point of view. 1

Abstract. We rely on the information from our trustworthy acquaintances to help us take even trivial decisions in our lives. Recommender Systems use the opinions of members of a community to help individuals in that community identify the information most likely to be interesting to them or relevant to their needs. These systems use the similarity between the user and recommenders or between the items to form recommendation list for the user. They do not take into consideration the social trust network between the entities in the society to ensure that the user can trust the recommendations received from the system. The paper proposes a model where a trust network exists between the peer agents and the personalized recommendations are generated on the basis of these trust relationships. The recommenders personalize recommendations by suggesting only those movies to user that matches its taste. Also, the social recommendation process is inherently fuzzy and uncertain. In the society, the information spreads through word-of-mouth and it is not possible to fully trust this information. There is uncertainty in the validity of such information. Again, when a product is recommended, it is suggested with linguistic quantifiers such as very good, more or less good, ordinary, and so on. Thus, uncertainty and fuzziness is inherent in the recommendation process. We have used Intuitionistic Fuzzy Sets to model such uncertainty and fuzziness in the recommendation process.

Abstract. Social networks can serve as both a rich source of new information and as a filter to identify the information most relevant to our specific needs. In this paper we present a methodology and algorithms that, by exploiting existing Semantic Web and Web2.0 data sources, help individuals identify who in their social network knows what, and who is the most trustworthy source of information on that topic. Our approach improves upon previous work in a number of ways, such as incorporating topic-specific rather than global trust metrics. This is achieved by generating topic experience profiles for each network member, based on data from Revyu and del.icio.us, to indicate who knows what. Identification of the most trustworthy sources is enabled by a rich trust model of information and recommendation seeking in social networks. Reviews and ratings created on Revyu provide source data for algorithms that generate topic expertise and person to person affinity metrics. Combining these metrics, we are implementing a user-oriented application for searching and automated ranking of information sources within social networks. 1