PROJECT TITLE :

Query Expansion with Enriched User Profiles for Personalized Search Utilizing Folksonomy Data - 2017

ABSTRACT:

Question expansion has been widely adopted in Web search as a approach of tackling the ambiguity of queries. Personalised search utilizing folksonomy information has demonstrated an extreme vocabulary mismatch problem that needs even more effective query enlargement ways. Co-prevalence statistics, tag-tag relationships, and semantic matching approaches are among those favored by previous research. However, user profiles which only contain a user's past annotation information may not be enough to support the selection of growth terms, especially for users with restricted previous activity with the system. We tend to propose a unique model to construct enriched user profiles with the help of an external corpus for customized question expansion. Our model integrates the present state-of-the-art text representation learning framework, known as word embeddings, with topic models in 2 groups of pseudo-aligned documents. Based mostly on user profiles, we build 2 novel question expansion techniques. These two techniques are primarily based on topical weights-enhanced word embeddings, and therefore the topical relevance between the query and also the terms inside a user profile, respectively. The results of an in-depth experimental evaluation, performed on two real-world datasets using completely different external corpora, show that our approach outperforms ancient techniques, as well as existing non-personalized and customized question growth methods.