Machine learning-based biomedical text classifying method and system and storage medium

Abstract

The invention discloses a text classifying method. The method comprises the steps of: inputting one or more texts; carrying out feature extraction on the one or more texts to obtain a first feature word set; at least inputting the first feature word set obtained by the feature extraction step to a machine learning algorithm to train and outputting a second feature word set; selecting feature wordsof the second feature word set to obtain a third feature word set; and inputting the third feature word set to the machine learning algorithm and training the machine learning algorithm again to obtain a classification algorithm; and classifying the one or more texts and / or one or more other additionally-input texts by using the classification algorithm. The invention also discloses a text classifying system and a computer-readable storage medium. A computer program is stored on the medium and the text classifying method is realized when the program is executed by a processor. Through the method disclosed by the invention, the selection of text features is improved, so that benefit is brought to improve the text classifying accuracy.

Description

[0001] The present invention claims the priority of Chinese Patent Application No. 201710266834.0, filed on April 21, 2017, entitled "Text Classification Method, System, and Computer-Readable Storage Medium", which is hereby incorporated by reference in its entirety. technical field [0002] The present invention generally relates to text classification methods, text classification systems, and computer-readable storage media. More specifically, it relates to a method and system for classifying medical literature texts and a computer-readable storage medium storing the text classification method. Background technique [0003] With the rapid development of sequencing technologies, various omics data have been easily generated and produced in large quantities, which is conducive to promoting the development of precision medicine. Typically, a knowledge database is a collection of new discoveries drawn from the literature and is a key component in interpreting data and translat...

AI Technical Summary

Problems solved by technology

However, the pace at which medical literature is being updated around the world increases the difficulty of text retrieval, especially for clinically meaningful information

However, due to the heterogeneity of data, the training method is a pure computer algorithm, and the artificial calibration and optimization of field experts are not included in the medical literature classification algorithm currently used in this technical field, the accuracy is low

The algorithm for text classification is not perfect, especially in the feature extraction of the text, which makes the marking of the text incomplete or inaccurate, and cannot accurately reflect the classification information of the text, thus making it difficult for subsequent retrieval work and database construction. Waiting brings a lot of distraction

Images