Web Mining and Web Usage Analysis 

Abstract: Significant vulnerabilities have recently been identified in collaborative filtering recommender systems. These vulnerabilities mostly emanate from the open nature of such systems and their reliance on user-specified judgments for building profiles. Attackers can easily introduce biased data in an attempt to force the system to “adapt” in a manner advantageous to them. Our research in secure personalization is examining a range of attack models, from the simple to the complex, and a variety of recommendation techniques. In this chapter, we explore an attack model that focuses on a subset of users with similar tastes and show that such an attack can be highly successful against both user-based and item-based collaborative filtering. We also introduce a detection model that can significantly decrease the impact of this attack.

Abstract: Personalized search has gained great popularity to improve search effectiveness in recent years. The objective of personalized search is to provide users with information tailored to their individual contexts. We propose to personalize Web search based on features extracted from hyperlinks, such as anchor terms or URL tokens. Our methodology personalizes PageRank vectors by weighting links based on the match between hyperlinks and user profiles. In particular, here we describe a profile representation using Internet domain features extracted from URLs. Users specify interest profiles as binary vectors where each feature corresponds to a set of one or more DNS tree nodes. Given a profile vector, a weighted PageRank is computed assigning a weight to each URL based on the match between the URL and the profile. We present promising results from an experiment in which users were allowed to select among nine URL features combining the top two levels of the DNS tree, leading to 29 pre-computed PageRank vectors from a Yahoo crawl. Personalized PageRank performed favorably compared to pure similarity based ranking and traditional PageRank.

Abstract: We present our work on automatically extracting social hierarchies from electronic communication data. Data mining based on user behavior can be leveraged to analyze and catalog patterns of communications between entities to rank relationships. The advantage is that the analysis can be done in an automatic fashion and can adopt itself to organizational changes over time. We illustrate the algorithms over real world data using the Enron corporation's email archive. The results show great promise when compared to the corporations work chart and judicial proceeding analyzing the major players.

Abstract: Recent studies have shown that conceptual and structural characteristics of a website can play an important role in the quality of recommendations provided by a recommendation system. Resources like Google Directory, Yahoo! Directory and web-content management systems attempt to organize content conceptually. Most recommendation models are limited in their ability to use this domain knowledge. We propose a novel technique to incorporate the conceptual characteristics of a website into a usage-based recommendation model. We use a framework based on biological sequence alignment. Similarity scores play a crucial role in such a construction and we introduce a scoring system that is generated from the website's concept hierarchy. These scores fit seamlessly with other quantities used in similarity calculation like browsing order and time spent on a page. Additionally they demonstrate a simple, extensible system for assimilating more domain knowledge. We provide experimental results to illustrate the benefits of using concept hierarchy.