Choices in knowledge-base retrieval for\ua0consumer health search

This paper investigates how retrieval using knowledge bases can be effectively translated to the consumer health search (CHS) domain. We posit that using knowledge bases for query reformulation may help to overcome some of the challenges in CHS. However, translating and implementing such approaches is nontrivial in CHS as it involves many design choices. We empirically evaluated the impact these different choices had on retrieval effectiveness. A state-of-the-art knowledge-base retrieval model—the Entity Query Feature Expansion model—was used to evaluate the following design choices: which knowledge base to use (specialised vs. generic), how to construct the knowledge base, how to extract entities from queries and map them to entities in the knowledge base, what part of the knowledge base to use for query expansion, and if to augment the KB search process with relevance feedback. While knowledge base retrieval has been proposed as a solution for CHS, this paper delves into the finer details of doing this effectively, highlighting both pitfalls and payoffs. It aims to provide some lessons to others in advancing the state-of-the-art in CHS