Rachael Rafter

Case-Based User Profiling for Content Personalisation

As it stands the Internets one size fits all approach to information retrieval presents the average user with a serious information overload problem. Adaptive hypermedia systems can provide a solution to this problem by learning about the implicit and explicit preferences of individual users and using this information to personalise information retrieval processes. We describe and evaluate a two-stage personalised information retrieval system that combines a server-side similarity-based retrieval component with a client-side case-based personalisation component. We argue that this combination has a number of benefits in terms of personalisation accuracy, computational cost, flexibility, security and privacy.

Automated Collaborative Filtering Applications for Online Recruitment Services

Online recruitment services suffer from shortcomings due to traditional search techniques. Most users fail to construct queries that provide an adequate and accurate description of their (job) requirements, leading to imprecise search results. We investigate one potential solution that combines implicit profiling methods and automated collaborative filtering (ACF) techniques to build personalised query-less job recommendations. Two ACF strategies are implemented and evaluated in the JobFinder domain.

Personalization techniques for online recruitment services

CASPER’s solution is a two-stage search engine (see the accompanying figure) that selects job cases not just according to their similarity to the target query, but also according to their relevance to the specific user in question, based on that user’s interaction history [1]. During stage one, job cases are ranked according to their similarity to the query by using a standard similarity metric, which calculates a similarity score between query features and corresponding job case features. Each case is made up of a fixed set of features such as job type, location, and salary (see [1]). This server-side stage produces a ranked list of job cases according to their similarity to the target query (see [2]). The second stage, a client-side process, emphasizes personalized information ordering. It reorders the results according to their relevance to the user by comparing each result to the user’s learned search profile. Each profile specifies the job cases the user has previously liked or disliked based on past feedback. Each stage-one result is associated with a relevance score by comparing it to the k most similar user profile cases. For instance, if the result is similar to many positive profile cases, it gets a high relevance score; whereas if it is similar to negative cases, it gets a low score. Thus priority is given to jobs that Online Recruitment Services Personalization Techniques for

Conversational Collaborative Recommendation --- An Experimental Analysis

Traditionally, collaborative recommender systems have been based on a single-shot model of recommendation where a single set of recommendations is generated based on a user’s (past) stored preferences. However, content-based recommender system research has begun to look towards more conversational models of recommendation, where the user is actively engaged in directing search at recommendation time. Such interactions can range from high-level dialogues with the user, possibly in natural language, to more simple interactions where the user is, for example, asked to indicate a preference for one of k suggested items. Importantly, the feedback attained from these interactions can help to differentiate between the user’s long-term stored preferences, and her current (short-term) requirements, which may be quite different. We argue that such interactions can also be beneficial to collaborative recommendation and provide experimental evidence to support this claim.

What Have the Neighbours Ever Done for Us? A Collaborative Filtering Perspective

Collaborative filtering (CF) techniques have proved to be a powerful and popular component of modern recommender systems. Common approaches such as user-based and item-based methods generate predictions from the past ratings of users by combining two separate ratings components: a base estimate , generally based on the average rating of the target user or item, and a neighbourhood estimate , generally based on the ratings of similar users or items. The common assumption is that the neighbourhood estimate gives CF techniques a considerable edge over simpler average-rating techniques. In this paper we examine this assumption more carefully and demonstrate that the influence of neighbours can be surprisingly minor in CF algorithms, and we show how this has been disguised by traditional approaches to evaluation, which, we argue, have limited progress in the field.

Great Explanations: Opinionated Explanations for Recommendations

Explaining recommendations helps users to make better decisions. We describe a novel approach to explanation for recommender systems, one that drives the recommendation ranking process, while at the same time providing the user with useful insights into the reason why items have been recommended and the trade-offs they may need to consider when making their choice. We describe this approach in the context of a case-based recommender system that harnesses opinions mined from user-generated reviews, and evaluate it on TripAdvisor hotel data.

The Recommendation Game: Using a Game-with-a-Purpose to Generate Recommendation Data

This paper describes a casual Facebook game to capture recommendation data as a side-effect of gameplay. We show how this data can be used to make successful recommendations as part of a live-user trial.

Opinionated Explanations for Recommendation Systems

This paper describes a novel approach for generating explanations for recommender systems based on opinions in user-generated reviews. We show how these opinions can be used to construct helpful and compelling explanations at recommendation time. The explanation highlights how the pros and cons of a recommended item compares to alternative items. We propose a way to score these explanations based on their content. The scores help to identify compelling explanations, providing a strong reason why the item being explained is better or worse than the alternatives. We describe the results of offline experiments and a live-user study based on TripAdvisor data to demonstrate the usefulness of this approach.

News Recommenders: Real-Time, Real-Life Experiences

In this paper we share our experiences of working with a real-time news recommendation framework with real-world user and data.

Recommending Topics for Web Curation

A new generation of curation services provides users with a set of tools to manually curate and manage topical collections of content. However, given curation is ultimately a manual effort, it still requires significant effort on the part of the curator both in terms of collecting and managing content. We are interested in providing additional assistance to users in their curation tasks, in particular when it comes to efficiently adding content to their collection, and examine recommender systems in an effort to automate this task. We examine a number of recommendation strategies using live-user data from the popular Scoop.it curation service.