{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/drug-drug-interaction-extraction-from","title":"Drug-Drug Interaction Extraction from Biomedical Text Using Long Short Term Memory Network","arxiv_id":"1701.08303","date":"2017-01-28","proceeding":null,"authors":["Sunil Kumar Sahu","Ashish Anand"],"abstract":"Simultaneous administration of multiple drugs can have synergistic or\nantagonistic effects as one drug can affect activities of other drugs.\nSynergistic effects lead to improved therapeutic outcomes, whereas,\nantagonistic effects can be life-threatening, may lead to increased healthcare\ncost, or may even cause death. Thus identification of unknown drug-drug\ninteraction (DDI) is an important concern for efficient and effective\nhealthcare. Although multiple resources for DDI exist, they are often unable to\nkeep pace with rich amount of information available in fast growing biomedical\ntexts. Most existing methods model DDI extraction from text as a classification\nproblem and mainly rely on handcrafted features. Some of these features further\ndepend on domain specific tools. Recently neural network models using latent\nfeatures have been shown to give similar or better performance than the other\nexisting models dependent on handcrafted features. In this paper, we present\nthree models namely, {\\it B-LSTM}, {\\it AB-LSTM} and {\\it Joint AB-LSTM} based\non long short-term memory (LSTM) network. All three models utilize word and\nposition embedding as latent features and thus do not rely on explicit feature\nengineering. Further use of bidirectional long short-term memory (Bi-LSTM)\nnetworks allow implicit feature extraction from the whole sentence. The two\nmodels, {\\it AB-LSTM} and {\\it Joint AB-LSTM} also use attentive pooling in the\noutput of Bi-LSTM layer to assign weights to features. Our experimental results\non the SemEval-2013 DDI extraction dataset show that the {\\it Joint AB-LSTM}\nmodel outperforms all the existing methods, including those relying on\nhandcrafted features. The other two proposed LSTM models also perform\ncompetitively with state-of-the-art methods.","url_abs":"http://arxiv.org/abs/1701.08303v2","url_pdf":"http://arxiv.org/pdf/1701.08303v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"drug-drug-interaction-extraction-from","repo_url":"https://github.com/sunilitggu/DDI-extraction-through-LSTM","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"drug-drug-interaction-extraction","task_name":"Drug–drug Interaction Extraction"},{"task_slug":"feature-engineering","task_name":"Feature Engineering"},{"task_slug":"medical-relation-extraction","task_name":"Medical Relation Extraction"},{"task_slug":"sentence","task_name":"Sentence"}],"methods":[{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}