Abstract
The BioCreative V chemical-disease relation (CDR) track was proposed to accelerate the progress of text mining in facilitating integrative understanding of chemicals, diseases and their relations. In this article, we describe an extension of our system (namely UET-CAM) that participated in the BioCreative V CDR. The original UET-CAM system’s performance was ranked fourth among 18 participating systems by the BioCreative CDR track committee. In the Disease Named Entity Recognition and Normalization (DNER) phase, our system employed joint inference (decoding) with a perceptron-based named entity recognizer (NER) and a back-off model with Semantic Supervised Indexing and Skip-gram for named entity normalization. In the chemical-induced disease (CID) relation extraction phase, we proposed a pipeline that includes a coreference resolution module and a Support Vector Machine relation extraction model. The former module utilized a multi-pass sieve to extend entity recall. In this article, the UET-CAM system was improved by adding a ‘silver’ CID corpus to train the prediction model. This silver standard corpus of more than 50 thousand sentences was automatically built based on the Comparative Toxicogenomics Database (CTD) database. We evaluated our method on the CDR test set. Results showed that our system could reach the state of the art performance with F1 of 82.44 for the DNER task and 58.90 for the CID task. Analysis demonstrated substantial benefits of both the multi-pass sieve coreference resolution method (F1 + 4.13%) and the silver CID corpus (F1 +7.3%).Database URL: SilverCID–The silver-standard corpus for CID relation extraction is freely online available at: https://zenodo.org/record/34530 (doi:10.5281/zenodo.34530).
Highlights
A survey of PubMed users’ search behavior showed that diseases and chemicals were two of the most frequently requested entities by PubMed users worldwide: diseases appeared in 20% of queries and chemicals in 11% [1]
The novel contributions of this article are as follows: (i) we proposed a Disease Named Entity Recognition and Normalization (DNER) model that was based on the joint inference between an averaged perceptron named entity recognizer (NER) model and a named entity normalization (NEN) pipeline of two phases, i.e. Supervised Semantic Indexing (SSI) followed by a skip-gram model; (ii) we demonstrated the benefit of our automatically built SilverCID corpus for the chemical-induced disease (CID) relation extraction; (iii) we presented evidence for the efficacy of using the multi-pass sieve in the CID relation extraction task and (iv) we demonstrated the strength of the rich feature set for CID relation extraction
Support Vector Machines (SVMs)-based intra-sentence relation extraction Our work was based on the know-how that if a noun phrase (NP) and an entity are coreferent, the NP can be considered as an entity of that type
Summary
A survey of PubMed users’ search behavior showed that diseases and chemicals were two of the most frequently requested entities by PubMed users worldwide: diseases appeared in 20% of queries and chemicals in 11% [1]. Chen et al [3] used this method to identify and rank associations between eight diseases and relevant drugs This approach tends to achieve high recall, but low precision and fails to distinguish the chemical-induced disease (CID) relations from other relations that commonly occur between chemicals and diseases. They, demands the time-consuming and labor-intensive manual compilation of huge knowledge (in terms of rules as in 4 or a three-tier hierarchical graph as in 5), which results from the wide variety of contexts in which relations can occur. These approaches, tend to suffer from the low recall. The variety of abundant ADR syntaxes as well as a failure to resolve intersentence alternative entity-mentions hampers the performance
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: Database : the journal of biological databases and curation
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
