A Generic Document Retrieval Framework Based on UMLS Similarity for Biomedical Question Answering System

Biomedical document retrieval systems play a vital role in biomedical question answering systems. The performance of the latter depends directly on the performance of its biomedical document retrieval section. Indeed, the main goal of biomedical document retrieval is to find a set of citations that have high probability to contain the answers. In this paper, we propose a biomedical document retrieval framework to retrieve the relevant documents for the biomedical questions (queries) from the users. In our framework, we first use GoPubMed search engine to find the top-K results. Then, we re-rank the top-K results by computing the semantic similarity between questions and the title of each document using UMLS similarity. Our proposed framework is evaluated on the BioASQ 2014 task datasets. The experimental results show that our proposed framework has the best performance (MAP@100) compared to the existing state-of-the-art related document retrieval systems.