BioNLP-OST 2019 RDoC Tasks: Multi-grain Neural Relevance Ranking Using Topics and Attention Based Query-Document-Sentence Interactions

TL;DR

This paper introduces neural relevance ranking methods using topics and attention mechanisms for retrieving and extracting relevant PubMed abstracts and sentences related to RDoC constructs, improving biomedical information retrieval.

Contribution

It proposes novel neural and relevance-based models for biomedical document retrieval and sentence extraction, achieving top performance in BioNLP-OST 2019 tasks.

Findings

Achieved 0.86 mAP in PubMed abstract retrieval

Secured 0.58 macro average accuracy in sentence extraction

Outperformed baseline models in relevance ranking

Abstract

This paper presents our system details and results of participation in the RDoC Tasks of BioNLP-OST 2019. Research Domain Criteria (RDoC) construct is a multi-dimensional and broad framework to describe mental health disorders by combining knowledge from genomics to behaviour. Non-availability of RDoC labelled dataset and tedious labelling process hinders the use of RDoC framework to reach its full potential in Biomedical research community and Healthcare industry. Therefore, Task-1 aims at retrieval and ranking of PubMed abstracts relevant to a given RDoC construct and Task-2 aims at extraction of the most relevant sentence from a given PubMed abstract. We investigate (1) attention based supervised neural topic model and SVM for retrieval and ranking of PubMed abstracts and, further utilize BM25 and other relevance measures for re-ranking, (2) supervised and unsupervised sentence ranking models utilizing multi-view representations comprising of query-aware attention-based sentence representation (QAR), bag-of-words (BoW) and TF-IDF. Our best systems achieved 1st rank and scored 0.86 mean average precision (mAP) and 0.58 macro average accuracy (MAA) in Task-1 and Task-2 respectively.