Format

Send to

Choose Destination
J Cheminform. 2018 May 23;10(1):28. doi: 10.1186/s13321-018-0280-0.

Putting hands to rest: efficient deep CNN-RNN architecture for chemical named entity recognition with no hand-crafted rules.

Author information

1
Laboratory of Functional Analysis of the Genome, Moscow Institute of Physics and Technology, Moscow, Russia. ilia.korvigo@gmail.com.
2
All-Russia Institute for Agricultural Microbiology, St. Petersburg, Russia. ilia.korvigo@gmail.com.
3
ITMO University, St. Petersburg, Russia. ilia.korvigo@gmail.com.
4
St. Petersburg State Pediatric Medical University, St. Petersburg, Russia.
5
N.N. Petrov Institute of Oncology, Department of Tumor Biology, St. Petersburg, Russia.
6
St. Petersburg State University, St. Petersburg, Russia.
7
Laboratory of Functional Analysis of the Genome, Moscow Institute of Physics and Technology, Moscow, Russia.
8
School of Biomedicine, Far Eastern Federal University, Vladivostok, Russia.
9
Laboratory of Functional Genomics, Research Centre for Medical Genetics, Moscow, Russia.

Abstract

Chemical named entity recognition (NER) is an active field of research in biomedical natural language processing. To facilitate the development of new and superior chemical NER systems, BioCreative released the CHEMDNER corpus, an extensive dataset of diverse manually annotated chemical entities. Most of the systems trained on the corpus rely on complicated hand-crafted rules or curated databases for data preprocessing, feature extraction and output post-processing, though modern machine learning algorithms, such as deep neural networks, can automatically design the rules with little to none human intervention. Here we explored this approach by experimenting with various deep learning architectures for targeted tokenisation and named entity recognition. Our final model, based on a combination of convolutional and stateful recurrent neural networks with attention-like loops and hybrid word- and character-level embeddings, reaches near human-level performance on the testing dataset with no manually asserted rules. To make our model easily accessible for standalone use and integration in third-party software, we've developed a Python package with a minimalistic user interface.

KEYWORDS:

Biocreative; Chemdner; Chemical; Conditional random fields; Convolutional neural network; Deep learning; Named entities recognition; Neural attention; Recurrent neural network; Text mining; Tokenisation

Supplemental Content

Full text links

Icon for Springer Icon for PubMed Central
Loading ...
Support Center