Joint Biomedical Entity and Relation Extraction with Knowledge-Enhanced Collective Inference

Abstract

Compared to the general news domain, information extraction (IE) from biomedical text requires much broader domain knowledge. However, many previous IE methods do not utilize any external knowledge during inference. Due to the exponential growth of biomedical publications, models that do not go beyond their fixed set of parameters will likely fall behind. Inspired by how humans look up relevant information to comprehend a scientific text, we present a novel framework that utilizes external knowledge for joint entity and relation extraction named KECI (Knowledge-Enhanced Collective Inference). Given an input text, KECI first constructs an initial span graph representing its initial understanding of the text. It then uses an entity linker to form a knowledge graph containing relevant background knowledge for the the entity mentions in the text. To make the final predictions, KECI fuses the initial span graph and the knowledge graph into a more refined graph using an attention mechanism. KECI takes a collective approach to link mention spans to entities by integrating global relational information into local representations using graph convolutional networks. Our experimental results show that the framework is highly effective, achieving new state-of-theart results in two different benchmark datasets: BioRelEx (binding interaction detection) and ADE (adverse drug event extraction). For example, KECI achieves absolute improvements of 4.59% and 4.91% in F1 scores over the stateof-the-art on the BioRelEx entity and relation extraction tasks 1 .

Artificial intelligence in recommender systems

“…Traditionally, natural language processing (NLP) has been developed with a focus on core tasks including translation and sentiment analysis. In the scientific domain, NLP tasks have focused on extracting information from the literature, such as named entity recognition [Li et al 2016d], entity linking [Lai et al 2021a], relation extraction Lai et al 2021b], and event extraction ]. NLP models have advanced rapidly in recent years and hence resulted in strong foundational models which can be easily applied to most NLP tasks [Devlin et al 2019;Raffel et al 2020;Brown et al 2020].…”

A Unified Transformer-based Framework for Duplex Text Normalization

“…With the rise of neural networks (NN) in natural language processing (NLP) [9,10,11], several deep learning models have recently been introduced for either TN or ITN [4,12,1,13]. For example, [14] uses recurrent NN-s to learn a TN function from a large corpus of written text aligned to its spoken form.…”

BERT might be Overkill: A Tiny but Effective Biomedical Entity Linker based on Residual Convolutional Neural Networks

“…", the mention NH3 should be linked to the entity KB:Ammonia. Biomedical EL is an important research problem, with applications in many downstream tasks, such as biomedical question answering , information retrieval, and information extraction (Wang et al, 2020;Lai et al, 2021b;. In general, two main challenges of the EL task are:…”