Drug interaction relationship extraction method and system based on external knowledge

Abstract

The invention provides a drug interaction relationship extraction method and system based on external knowledge, and the method comprises the following steps: carrying out the analysis and processingof the content of a drug database, extracting and generating interacted drug pairs, storing all drug description information, and forming a drug interaction data set with the drug description information; constructing a drug description system information training model, and performing training through the drug interaction data set to obtain and store an optimal model; combining the optimal modelwith a BiLSTM-Att-CapsNet model, so that an EK-BiLSTM-Att-CapsNet model can be obtained, meanwhile, drug entities of the drug interaction data set are recognized, corresponding drug description information is found in a drug database and stored, and finally, the combined model is trained to obtain a final relationship extraction model. According to the method, the problem of large difference of extraction results of different relationship categories can be relieved, and the extraction effect is improved.

Description

technical field [0001] The invention belongs to the technical field of natural language processing, and in particular relates to a method for extracting drug interaction relations based on external knowledge. Background technique [0002] Drug-Drug Interactions (DDI) refers to the synergistic or antagonistic effects between different drugs when patients take multiple drugs at the same time, which may cause side effects, resulting in increased treatment costs and harm to patients. Therefore, understanding the interaction knowledge between drugs is of great significance and value for the diagnosis and treatment of patients and the development of medicine. [0003] At present, the application methods in the field of drug interaction relationship extraction mainly include: rule-based methods, traditional machine learning-based methods and deep learning-based methods. In the rule-based method, the formulation of the rules generally requires the assistance of professionals in the...

AI Technical Summary

Problems solved by technology

In the rule-based method, the formulation of the rules generally requires the assistance of professionals in the medical field. Due to the diversity of language expressions, the formulated rules are often difficult to cover all drug interaction relationships, so the recall rate of this method is low; Traditional machine learning methods usually need to use a large number of manually defined features, such as parts of speech, syntax, grammar, etc., and need to use external natural language processing tools to generate these features, such as part-of-speech taggers, syntactic analyzers, etc., so its extraction performance It is greatly affected by external natural language processing tools; the method based on deep learning has the ability to automatically learn features, which can reduce the cost of manual design features and the extraction effect is generally better than traditional methods, but similar to the previous two methods, There will be large differences in the extraction results of the model on different relationship categories

Images