Recognizing Mentions of Adverse Drug Reaction in Social Media Using
            Knowledge-Infused Recurrent Models

Stanovsky, Gabriel; Gruhl, Daniel; Mendes, Pablo N.

doi:10.18653/v1/e17-1014

Cited by 48 publications

(39 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…can help [47], it makes sense to consider the "best order" to ask the user for input in the hopes of achieving a sufficiently performant system with minimal human effort.…”

Section: Methodsmentioning

confidence: 99%

Exploring the Efficiency of Batch Active Learning for Human-in-the-Loop Relation Extraction

Lourentzou

Gruhl

Welch

2018

Companion of the the Web Conference 2018 on the Web Conference 2018 - WWW '18

Self Cite

View full text Add to dashboard Cite

Domain-specific relation extraction requires training data for supervised learning models, and thus, significant labeling effort. Distant supervision is often leveraged for creating large annotated corpora however these methods require handling the inherent noise. On the other hand, active learning approaches can reduce the annotation cost by selecting the most beneficial examples to label in order to learn a good model. The choice of examples can be performed sequentially, i.e. select one example in each iteration, or in batches, i.e. select a set of examples in each iteration. The optimization of the batch size is a practical problem faced in every real-world application of active learning, however it is often treated as a parameter decided in advance. In this work, we study the trade-off between model performance, the number of requested labels in a batch and the time spent in each round for real-time, domain specific relation extraction. Our results show that the use of an appropriate batch size produces competitive performance, even compared to a fully sequential strategy, while reducing the training time dramatically.

show abstract

“…can help [47], it makes sense to consider the "best order" to ask the user for input in the hopes of achieving a sufficiently performant system with minimal human effort.…”

Section: Methodsmentioning

confidence: 99%

Exploring the Efficiency of Batch Active Learning for Human-in-the-Loop Relation Extraction

Lourentzou

Gruhl

Welch

2018

Companion of the the Web Conference 2018 on the Web Conference 2018 - WWW '18

Self Cite

View full text Add to dashboard Cite

show abstract

“…Recall F1-score Baseline [5] 0.7067 ± 0.057 0.7207 ± 0.074 0.7102 ± 0.049 Baseline with adam 0.7065 ± 0.058 0.7576 ± 0.083 0.7272 ± 0.051 KB-Embedding Baseline [22] 0.7171 ± 0.058 0.7713 ± 0.091 0.7397 ± 0.055 Self-training 0.6999 ± 0.047 0.8304 ± 0.039 0.7588 ± 0.039 Joint MTL (Section 3.3) 0.7177 ± 0.027 0.8482 ± 0.068 0.7770 ± 0.043 MTL (Section 3.2) 0.7569 ± 0.044 0.8386 ± 0.078 0.7935 ± 0.045 set to 64 with the confidence threshold value empirically set to 0.5. The stopping criteria for the self-training kicks in when the number of iterations reaches 5 or if the unlabeled tweets pool is exhausted, whichever occurs first.…”

Section: Methods Precisionmentioning

confidence: 99%

“…Precision Recall F1-score Baseline [5] 0.6120 ± 0.116 0.5149 ± 0.099 0.5601 ± 0.100 Baseline with adam 0.6281 ± 0.094 0.5614 ± 0.110 0.5859 ± 0.079 KB-Embedding Baseline [22] 0.5960 ± 0.081 0.6144 ± 0.068 0.6042 ± 0.060 Self-training 0.5717 ± 0.056 0.7141 ± 0.082 0.6332 ± 0.057 Joint MTL (Section 3. 3) 0.5675 ± 0.049 0.7384 ± 0.079 0.6401 ± 0.051 MTL (Section 3.2) 0.6656 ± 0.083 0.6380 ± 0.077 0.6482 ± 0.065 The KB-embedding baseline [22] replaces word embeddings of the medical entities in the sentence with the corresponding embeddings learned from a knowledge-base. The corresponding results can be seen in row 3 of the tables.…”

Section: Methodsmentioning

confidence: 99%

“…Cocos et al [5] proposed a Long Short Term Memory (LSTM) based model with word embedding features to extract ADRs from Twitter posts. Stanovsky et al [22] proposed a LSTM based model where lexical word embeddings are augmented with Knowledge-Graph based embeddings. In their model, if a word has a lexical match with a Knowledge-Graph entity (e.g., DBPedia), its corresponding lexical word embedding is replaced by embedding learned through Knowledge graph based methods [25].…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Multi-task Learning for Extraction of Adverse Drug Reaction Mentions from Tweets

Gupta

Varma

et al. 2018

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Adverse drug reactions (ADRs) are one of the leading causes of mortality in health care. Current ADR surveillance systems are often associated with a substantial time lag before such events are officially published. On the other hand, online social media such as Twitter contain information about ADR events in real-time, much before any official reporting. Current state-of-the-art in ADR mention extraction uses Recurrent Neural Networks (RNN), which typically need large labeled corpora. Towards this end, we propose a multi-task learning based method which can utilize a similar auxiliary task (adverse drug event detection) to enhance the performance of the main task, i.e., ADR extraction. Furthermore, in absence of the auxiliary task dataset, we propose a novel joint multi-task learning method to automatically generate weak supervision dataset for the auxiliary task when a large pool of unlabeled tweets is available. Experiments with ∼0.48M tweets show that the proposed approach outperforms the state-of-the-art methods for the ADR mention extraction task by ∼7.2% in terms of F1 score.

show abstract

“…Deep learning methods have been increasingly applied to solve NLP tasks in the medical domain and achieve better performance than CRFs. In the case of ADR mention recognition, Stanovsky et al [40] employed RNNs with word embeddings trained on a Blekko medical corpus in conjunction with entity embeddings trained on DBpedia. If an entity mention was a lexical match with one of DBpedia entities, then the entity embeddings trained on DBpedia replaced word embeddings of all words in the entity mention.…”

Section: Related Workmentioning

confidence: 99%

Recognizing Continuous and Discontinuous Adverse Drug Reaction Mentions from Social Media Using LSTM‐CRF

Tang

Wang

et al. 2018

Wireless Communications and Mobile Computing

View full text Add to dashboard Cite

Social media in medicine, where patients can express their personal treatment experiences by personal computers and mobile devices, usually contains plenty of useful medical information, such as adverse drug reactions (ADRs); mining this useful medical information from social media has attracted more and more attention from researchers. In this study, we propose a deep neural network (called LSTM-CRF) combining long short-term memory (LSTM) neural networks (a type of recurrent neural networks) and conditional random fields (CRFs) to recognize ADR mentions from social media in medicine and investigate the effects of three factors on ADR mention recognition. The three factors are as follows: (1) representation for continuous and discontinuous ADR mentions: two novel representations, that is, "BIOHD" and "Multilabel," are compared; (2) subject of posts: each post has a subject (i.e., drug here); and (3) external knowledge bases. Experiments conducted on a benchmark corpus, that is, CADEC, show that LSTM-CRF achieves better -score than CRF; "Multilabel" is better in representing continuous and discontinuous ADR mentions than "BIOHD"; both subjects of comments and external knowledge bases are individually beneficial to ADR mention recognition. To the best of our knowledge, this is the first time to investigate deep neural networks to mine continuous and discontinuous ADRs from social media.

show abstract

Recognizing Mentions of Adverse Drug Reaction in Social Media Using Knowledge-Infused Recurrent Models

Cited by 48 publications

References 23 publications

Exploring the Efficiency of Batch Active Learning for Human-in-the-Loop Relation Extraction

Exploring the Efficiency of Batch Active Learning for Human-in-the-Loop Relation Extraction

Multi-task Learning for Extraction of Adverse Drug Reaction Mentions from Tweets

Recognizing Continuous and Discontinuous Adverse Drug Reaction Mentions from Social Media Using LSTM‐CRF

Contact Info

Product

Resources

About