Now showing 1 - 10 of 53
  • Publication
    Towards the Recommendation of Personalised Activity Sequences in the Tourism Domain
    In this paper we consider the problem of recommending sequencesof activities to a user. The proposed approach leverages the order aswell as the context associated with the users past activity patternsto make recommendations. This work extends the general activityrecommendation framework proposed in [16] to iteratively recommendthe next sequence of activities to perform. We demonstratethe efficacy of our recommendation framework by applying it to thetourism domain and evaluations are performed using a real-world(checkin) dataset
  • Publication
    A multi-criteria evaluation of a user generated content based recommender system
    The Social Web provides new and exciting sources of information that may be used by recommender systems as a complementary source of recommendation knowledge. For example, User-Generated Content, such as reviews, tags, comments, tweets etc. can provide a useful source of item information and user preference data, if a clear signal can be extracted from the inevitable noise that exists within these sources. In previous work we explored this idea, mining term-based recommendation knowledge from user reviews, to develop a recommender that compares favourably to conventional collaborative-filtering style techniques across a range of product types. However, this previous work focused solely on recommendation accuracy and it is now well accepted in the literature that accuracy alone tells just part of the recommendation story. For example, for many, the promise of recommender systems lies in their ability to surprise with novel recommendations for less popular items that users might otherwise miss. This makes for a riskier recommendation prospect, of course, but it could greatly enhance the practical value of recommender systems to end-users. In this paper we analyse our User-Generated Content (UGC) approach to recommendation using metrics such as novelty, diversity, and coverage and demonstrate superior performance, when compared to conventional user-based and item- based collaborative filtering techniques, while highlighting a number of interesting performance trade-offs.
  • Publication
    A recommender system approach to enhance web search and query formulation
    While search engines are the primary means by which information is located online, significant issues remain when trying to satisfy the needs of searchers, especially in the face of the type of vague queries that dominate Web search. In this paper, we tackle this problem by applying a recommender system approach to Web search which allows users to dynamically interact with the result-space that is of interest to them. Our proposed recommendation interface also facilitates query expansion through a context-sensitive tag cloud, helping searchers to efficiently assimilate potential expansion terms that are mined from results of interest. We present findings from a live user trial of our approach which indicate, for example, that it facilitates users to locate relevant information more quickly when compared to using standard search engine result lists.
  • Publication
    Unsupervised Retrieval of Attack Profiles in Collaborative Recommender Systems
    (University College Dublin. School of Computer Science and Informatics, 2008-04) ; ;
    Trust, reputation and recommendation are key components of successful ecommerce systems. However, ecommerce systems are also vulnerable in this respect because there are opportunities for sellers to gain advantage through manipulation of reputation and recommendation. One such vulnerability is the use of fraudulent user profiles to boost (or damage) the ratings of items in an online recommender system. In this paper we cast this problem as a problem of detecting anomalous structure in network analysis and propose a novel mechanism for detecting this anomalous structure. We present an evaluation that shows that this approach is effective at uncovering the types of recommender systems attack described in the literature.
  • Publication
    Collaboration and reputation in social web search
    Recent research has highlighted the inherently collaborative nature of many Web search tasks, even though collaborative searching is not supported by mainstream search engines. In this paper, we examine the activity of early adopters of HeyStaks, a collaborative Web search framework that is designed to complement mainstream search engines such as Google, Bing, and Yahoo. The utility allows users to search as normal, using their favourite search engine, while benefiting from a more collaborative and social search experience. HeyStaks supports searchers by harnessing the experiences of others, in order to enhance organic mainstream result-lists. We review some early evaluation results that speak to the practical benefits of search collaboration in the context of the recently proposed Reader-to-Leader social media analysis framework [11]. In addition, we explore the idea of utilising the reputation model introduced by McNally et al.[6] in order to identify the search leaders in HeyStaks, i.e. those users who are responsible for driving collaboration in the HeyStaks application.
  • Publication
    Evaluating user reputation in collaborative web search
    Often today’s recommender systems look to past user activity in order to influence future recommendations. In the case of social web search, employing collaborative recommendation techniques allows for personalization of search results. If recommendations arise from past user activity, the expertise of those users driving the recommendation process can play an important role when it comes to ensuring recommendation quality. Hence the reputation of users is important in collaborative and social search tasks, in addition to result relevance as traditionally considered in web search. In this paper we explore this concept of reputation; specifically, investigating how reputation can enhance the recommendation engine at the core of the HeyStaks social search utility. We evaluate a number of different reputation models in the context of the HeyStaks system, and demonstrate how incorporating reputation into the recommendation process can enhance the relevance of results recommended by HeyStaks.
  • Publication
    A Model of Collaboration-based Reputation for the Social Web
    In this paper we describe a generic approach to modeling user reputation in online social platforms based on an underlying model of collaboration. This distinguishes our approach from more conventional reputation models which are often based around ad-hoc activity metrics. We evaluate our model with respect to a conventional reputation model used by 3 social Q&A websites, each based on a different topical domain.
  • Publication
    Further experiments in micro-blog categorization
    (Intelligent Systems Research Centre, 2011-08-31) ; ;
    Since the creation of Twitter in 2008, micro-blogging services have received a lot of attention among users who wish to share news items, opinions and information with friends and colleagues. However, these services typically provide for only limited organisation of content, with the main ranking criterion being post time with perhaps some basic message filtering accommodated. Given the substantial and increasing volume of posts that micro-blogging services attract, there is a clear need to assist users when it comes to effectively consuming this content. In this regard, categorisation offers one approach to organise content by grouping related messages together. In this paper we present a study in the recommendation of categories for short-form messages in order to provide for better search and message filtering. In particular, we present an index-based approach where real-time web data can be used as a source of knowledge for category recommendation. Further, we evaluate our approach on two different micro-blogging datasets and results show that micro-blog messages in sufficient quantities provide a useful recommendation signal for category recommendation.
  • Publication
    The readability of helpful product reviews
    Consumers frequently rely on user-generated product reviews to guide purchasing decisions. Given the ever increasing volume of such reviews and variations in review quality, consumers require assistance to effectively leverage this vast information source. In this paper, we examine to what extent the readability of reviews is a predictor of review helpfulness. Using a supervised classification approach, our findings indicate that readability is a useful predictor for Amazon product reviews but less so for TripAdvisor hotel reviews.
  • Publication
    Evaluating the Relative Performance of Neighbourhood-Based Recommender Systems
    Neighbourhood-based recommender systems are a class of collaborative filtering algorithms, which rely on finding like-minded users to generate recommendations, automating what is usually known as word-of-mouth. These systems attempt to solve the information overload problem by presenting the user with relevant items. However, there is evidence showing that these algorithms may contribute to the filter bubble problem, making it harder for the user to find interesting items which are non-popular. In this paper we propose a novel evaluation of the performance and biases of the two most common neighbourhood-based approaches: user k-nearest neighbour collaborative filtering (UKNN ), and item k-nearest neighbour collaborative filtering (IKNN). We propose an evaluation which considers the size of the neighbourhood, finding that optimising for accuracy in UKNN algorithms leads to a poor performance in terms of diversity, a higher bias towards popularity, and less unique recommendations, when compared to the IKNN approach.