Interactive Attention Network for Adverse Drug Reaction Classification

Alimova, Ilseyar; Solovyev, Valery D.

doi:10.1007/978-3-030-01204-5_18

Cited by 10 publications

(7 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If the relationship between adverse reactions and drugs was labeled as Outcome-negative in the sentence, we marked the sentence as ADR (positive), otherwise, we annotate it as non-ADR (negative). The data division was similar to that used in Ilseyar et al [24]. • ADE [8].…”

Section: Experimental Datasets and Settingsmentioning

confidence: 99%

“…In the first two lines of Table 2, we assess the performance of the main model and baseline that was proposed by Alimova et al [24] The feature-rich SVM method is based on the SVM with a linear kernel [41]. It considered a set of features.…”

Section: Evaluation On Twimedmentioning

confidence: 99%

“…However, in our method, we can still get better results with a few features. In the second line, the method utilized an interactive attention network (IAN) [24] to learn the representations for targets and contexts. The IAN used attention mechanisms to detect the important words of the target expression and its full context.…”

Section: Evaluation On Twimedmentioning

confidence: 99%

“…These experimental results suggest that our MSAM model could combine the contextual features that are extracted by Bi-LSTM. Compared with the feature-rich SVM [24] method, our method effectively reduces feature construction. We apply multiple self-attention steps to learn the representations for sentences.…”

Section: Evaluation On Twimedmentioning

confidence: 99%

“…The segments with a stronger focus in the sentence are treated as more important, which would influence the sentence representation. Alimova et al [24] investigated the applicability of the interactive attention network (IAN) for the identification of adverse drug reactions from user reviews. Lin et al [25] and Ji et al [26] introduced an attention mechanism to the PCNN-based multi-instance learning (MIL) framework to select informative sentences.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Adverse drug reaction detection via a multihop self-attention mechanism

et al. 2019

View full text Add to dashboard Cite

Background: The adverse reactions that are caused by drugs are potentially life-threatening problems. Comprehensive knowledge of adverse drug reactions (ADRs) can reduce their detrimental impacts on patients. Detecting ADRs through clinical trials takes a large number of experiments and a long period of time. With the growing amount of unstructured textual data, such as biomedical literature and electronic records, detecting ADRs in the available unstructured data has important implications for ADR research. Most of the neural network-based methods typically focus on the simple semantic information of sentence sequences; however, the relationship of the two entities depends on more complex semantic information. Methods: In this paper, we propose multihop self-attention mechanism (MSAM) model that aims to learn the multi-aspect semantic information for the ADR detection task. first, the contextual information of the sentence is captured by using the bidirectional long short-term memory (Bi-LSTM) model. Then, via applying the multiple steps of an attention mechanism, multiple semantic representations of a sentence are generated. Each attention step obtains a different attention distribution focusing on the different segments of the sentence. Meanwhile, our model locates and enhances various keywords from the multiple representations of a sentence. Results: Our model was evaluated by using two ADR corpora. It is shown that the method has a stable generalization ability. Via extensive experiments, our model achieved F-measure of 0.853, 0.799 and 0.851 for ADR detection for TwiMed-PubMed, TwiMed-Twitter, and ADE, respectively. The experimental results showed that our model significantly outperforms other compared models for ADR detection. Conclusions: In this paper, we propose a modification of multihop self-attention mechanism (MSAM) model for an ADR detection task. The proposed method significantly improved the learning of the complex semantic information of sentences.

show abstract

Section: Experimental Datasets and Settingsmentioning

confidence: 99%

Section: Evaluation On Twimedmentioning

confidence: 99%

Section: Evaluation On Twimedmentioning

confidence: 99%

Section: Evaluation On Twimedmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations