Dataset Information

A novel framework for biomedical entity sense induction.

ABSTRACT: BACKGROUND:Rapid advancements in biomedical research have accelerated the number of relevant electronic documents published online, ranging from scholarly articles to news, blogs, and user-generated social media content. Nevertheless, the vast amount of this information is poorly organized, making it difficult to navigate. Emerging technologies such as ontologies and knowledge bases (KBs) could help organize and track the information associated with biomedical research developments. A major challenge in the automatic construction of ontologies and KBs is the identification of words with its respective sense(s) from a free-text corpus. Word-sense induction (WSI) is a task to automatically induce the different senses of a target word in the different contexts. In the last two decades, there have been several efforts on WSI. However, few methods are effective in biomedicine and life sciences. METHODS:We developed a framework for biomedical entity sense induction using a mixture of natural language processing, supervised, and unsupervised learning methods with promising results. It is composed of three main steps: (1) a polysemy detection method to determine if a biomedical entity has many possible meanings; (2) a clustering quality index-based approach to predict the number of senses for the biomedical entity; and (3) a method to induce the concept(s) (i.e., senses) of the biomedical entity in a given context. RESULTS:To evaluate our framework, we used the well-known MSH WSD polysemic dataset that contains 203 annotated ambiguous biomedical entities, where each entity is linked to 2-5 concepts. Our polysemy detection method obtained an F-measure of 98%. Second, our approach for predicting the number of senses achieved an F-measure of 93%. Finally, we induced the concepts of the biomedical entities based on a clustering algorithm and then extracted the keywords of reach cluster to represent the concept. CONCLUSIONS:We have developed a framework for biomedical entity sense induction with promising results. Our study results can benefit a number of downstream applications, for example, help to resolve concept ambiguities when building Semantic Web KBs from biomedical text.

SUBMITTER: Lossio-Ventura JA

PROVIDER: S-EPMC6766751 | biostudies-literature | 2018 Aug

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A novel framework for biomedical entity sense induction.

Lossio-Ventura J A JA Bian J J Jonquet C C Roche M M Teisseire M M

Journal of biomedical informatics 20180620

<h4>Background</h4>Rapid advancements in biomedical research have accelerated the number of relevant electronic documents published online, ranging from scholarly articles to news, blogs, and user-generated social media content. Nevertheless, the vast amount of this information is poorly organized, making it difficult to navigate. Emerging technologies such as ontologies and knowledge bases (KBs) could help organize and track the information associated with biomedical research developments. A ma ...[more]

PMID: 29935347

Similar Datasets

Project description:MotivationThe explosive increase of biomedical literature has made information extraction an increasingly important tool for biomedical research. A fundamental task is the recognition of biomedical named entities in text (BNER) such as genes/proteins, diseases and species. Recently, a domain-independent method based on deep learning and statistical word embeddings, called long short-term memory network-conditional random field (LSTM-CRF), has been shown to outperform state-of-the-art entity-specific BNER tools. However, this method is dependent on gold-standard corpora (GSCs) consisting of hand-labeled entities, which tend to be small but highly reliable. An alternative to GSCs are silver-standard corpora (SSCs), which are generated by harmonizing the annotations made by several automatic annotation systems. SSCs typically contain more noise than GSCs but have the advantage of containing many more training examples. Ideally, these corpora could be combined to achieve the benefits of both, which is an opportunity for transfer learning. In this work, we analyze to what extent transfer learning improves upon state-of-the-art results for BNER.ResultsWe demonstrate that transferring a deep neural network (DNN) trained on a large, noisy SSC to a smaller, but more reliable GSC significantly improves upon state-of-the-art results for BNER. Compared to a state-of-the-art baseline evaluated on 23 GSCs covering four different entity classes, transfer learning results in an average reduction in error of approximately 11%. We found transfer learning to be especially beneficial for target datasets with a small number of labels (approximately 6000 or less).Availability and implementationSource code for the LSTM-CRF is available at https://github.com/Franck-Dernoncourt/NeuroNER/ and links to the corpora are available at https://github.com/BaderLab/Transfer-Learning-BNER-Bioinformatics-2018/.Supplementary informationSupplementary data are available at Bioinformatics online.

Project description:Increased interest in the use of word embeddings, such as word representation, for biomedical named entity recognition (BioNER) has highlighted the need for evaluations that aid in selecting the best word embedding to be used. One common criterion for selecting a word embedding is the type of source from which it is generated; that is, general (e.g., Wikipedia, Common Crawl), or specific (e.g., biomedical literature). Using specific word embeddings for the BioNER task has been strongly recommended, considering that they have provided better coverage and semantic relationships among medical entities. To the best of our knowledge, most studies have focused on improving BioNER task performance by, on the one hand, combining several features extracted from the text (for instance, linguistic, morphological, character embedding, and word embedding itself) and, on the other, testing several state-of-the-art named entity recognition algorithms. The latter, however, do not pay great attention to the influence of the word embeddings, and do not facilitate observing their real impact on the BioNER task. For this reason, the present study evaluates three well-known NER algorithms (CRF, BiLSTM, BiLSTM-CRF) with respect to two corpora (DrugBank and MedLine) using two classic word embeddings, GloVe Common Crawl (of the general type) and Pyysalo PM + PMC (specific), as unique features. Furthermore, three contextualized word embeddings (ELMo, Pooled Flair, and Transformer) are compared in their general and specific versions. The aim is to determine whether general embeddings can perform better than specialized ones on the BioNER task. To this end, four experiments were designed. In the first, we set out to identify the combination of classic word embedding, NER algorithm, and corpus that results in the best performance. The second evaluated the effect of the size of the corpus on performance. The third assessed the semantic cohesiveness of the classic word embeddings and their correlation with respect to several gold standards; while the fourth evaluates the performance of general and specific contextualized word embeddings on the BioNER task. Results show that the classic general word embedding GloVe Common Crawl performed better in the DrugBank corpus, despite having less word coverage and a lower internal semantic relationship than the classic specific word embedding, Pyysalo PM + PMC; while in the contextualized word embeddings the best results are presented in the specific ones. We conclude, therefore, when using classic word embeddings as features on the BioNER task, the general ones could be considered a good option. On the other hand, when using contextualized word embeddings, the specific ones are the best option.

Dataset Information

A novel framework for biomedical entity sense induction.

Publications

A novel framework for biomedical entity sense induction.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets