NLNDE: Enhancing Neural Sequence Taggers with Attention and Noisy Channel for Robust Pharmacological Entity Detection

TL;DR

This paper presents NLNDE, a robust neural sequence tagging system for pharmacological entity detection in Spanish texts, leveraging attention mechanisms and noisy channel models to improve performance without domain or language expertise.

Contribution

The paper introduces a novel architecture combining attention-based embedding selection and automatic data annotation for domain- and language-agnostic pharmacological entity recognition.

Findings

Achieved up to 88.6% F1 score in PharmaCoNER competition

Demonstrated effectiveness of combining attention and noisy channel techniques

System performs well without requiring domain or language-specific knowledge

Abstract

Named entity recognition has been extensively studied on English news texts. However, the transfer to other domains and languages is still a challenging problem. In this paper, we describe the system with which we participated in the first subtrack of the PharmaCoNER competition of the BioNLP Open Shared Tasks 2019. Aiming at pharmacological entity detection in Spanish texts, the task provides a non-standard domain and language setting. However, we propose an architecture that requires neither language nor domain expertise. We treat the task as a sequence labeling task and experiment with attention-based embedding selection and the training on automatically annotated data to further improve our system's performance. Our system achieves promising results, especially by combining the different techniques, and reaches up to 88.6% F1 in the competition.