Exploring Recommenders for Improved Information Retrieval

The continuing increase in the volume of information available in our daily lives is creating ever greater challenges for people to find personally useful information. One approach used to addressing this challenge is Personalized Information Retrieval (PIR) which develops personal user profiles with the objective of giving retrieval results which better meet individual user information needs. However, in many situations there may be no opportunity to build a personal profile, e.g. when this user is querying on a new topic. Under this condition, the experiences and behaviour of other previous users, who have made similar queries, could be used to build a model of general user behaviour in this domain. Our research proposes to focus on the development a new IR approach for topic domains. Our work seeks to combine recommender algorithms trained using previous search behaviour from different searchers with a standard ranked IR method to form a topic-specific IR model to improve the search effectiveness for a user entering a query without personal prior search history on this topic. The challenges of our work are: how to provide users better results; what data collection can be used to conduct experiment; and how to do the combination of recommender and IR components.