An attention-based effective neural model for drug-drug interactions extraction

Abstract

BackgroundDrug-drug interactions (DDIs) often bring unexpected side effects. The clinical recognition of DDIs is a crucial issue for both patient safety and healthcare cost control. However, although text-mining-based systems explore various methods to classify DDIs, the classification performance with regard to DDIs in long and complex sentences is still unsatisfactory.MethodsIn this study, we propose an effective model that classifies DDIs from the literature by combining an attention mechanism and a recurrent neural network with long short-term memory (LSTM) units. In our approach, first, a candidate-drug-oriented input attention acting on word-embedding vectors automatically learns which words are more influential for a given drug pair. Next, the inputs merging the position- and POS-embedding vectors are passed to a bidirectional LSTM layer whose outputs at the last time step represent the high-level semantic information of the whole sentence. Finally, a softmax layer performs DDI classification.ResultsExperimental results from the DDIExtraction 2013 corpus show that our system performs the best with respect to detection and classification (84.0% and 77.3%, respectively) compared with other state-of-the-art methods. In particular, for the Medline-2013 dataset with long and complex sentences, our F-score far exceeds those of top-ranking systems by 12.6%.ConclusionsOur approach effectively improves the performance of DDI classification tasks. Experimental analysis demonstrates that our model performs better with respect to recognizing not only close-range but also long-range patterns among words, especially for long, complex and compound sentences.

Identifying Common Methods Used by Drug Interaction Experts for Finding Evidence About Potential Drug-Drug Interactions: Web-Based Survey (Preprint)

“…Other studies have focused on data mining to identify PDDIs within titles, abstracts, and articles [16][17][18][19][20][21][22][23][24][25][26][27][28]. Included in these studies are approaches to use various kinds of machine learning including linear kernels (e.g., Support Vector Machines) [18,19,32], non-linear kernels (e.g., Graph Models) [22], random forest [16], various neural network architectures [17,21,26,33], advanced use of linguistic, parts of speech and linguistic features [19,23], unsupervised topical models [25], and semantic features from terminologies or ontologies [16,27,32,35].…”

“…Included in these studies are approaches to use various kinds of machine learning including linear kernels (e.g., Support Vector Machines) [18,19,32], non-linear kernels (e.g., Graph Models) [22], random forest [16], various neural network architectures [17,21,26,33], advanced use of linguistic, parts of speech and linguistic features [19,23], unsupervised topical models [25], and semantic features from terminologies or ontologies [16,27,32,35]. In general, the goal of these sophisticated approaches is to accurately extract PDDI data from the large body of scientific literature.…”

Relation Extraction Between Bacteria and Biotopes from Biomedical Texts with Attention Mechanisms and Domain-Specific Contextual Representations

“…Attention mechanisms have recently been successfully applied to biomedical relation extraction tasks [14,18,30]. These attention networks are able to learn a vector of important weights for each word in a sentence to reflect its impact on the final result.…”

“…However, it is very difficult for DL models to learn enough features from only sequences of sentences. Instead of learning from full sentences, attention networks have demonstrated success in a wide range of NLP tasks [25,26,27,28,29,30,31]. In addition, BGRU-Attn [18] first used the Additive attention mechanism [29] for the BB task to focus on only sections of the output from RNN instead of the entire outputs and achieved state-of-the-art performance.…”

“…In addition, BGRU-Attn [18] first used the Additive attention mechanism [29] for the BB task to focus on only sections of the output from RNN instead of the entire outputs and achieved state-of-the-art performance. Other attention techniques such as Entity-Oriented attention [30] and Multi-Head attention [31] still have not been explored for this task. From the aspect of word representation, traditional word embeddings [32,33] only allow for single contextindependent representation.…”