Chinese short text subjective question automatic scoring method and system using LSTM neural network

Abstract

The invention provides a Chinese short text subjective question automatic scoring method using an LSTM neural network. The method comprises the steps that firstly, an answer text is segmented, and thetext is converted into a word sequence; secondly, a vectorization expression of each word in the answer text is obtained, and an answer text mapping matrix is constructed; thirdly, the LSTM neural network is used for carrying out operation on the answer text mapping matrix, output of all or a part of hidden layers is obtained to obtain a semantic feature matrix of the answer text; fourthly, down-sampling is conducted on the semantic feature matrix by utilizing a pooling algorithm to obtain a semantic feature vector of the answer text; fifthly, the semantic feature vector of the answer text isgiven to a classifier, and the category of the answer text is predicted; sixthly, the many-to-one relationship between the category where the answer text belongs and the score is considered, and thescore of the answer text is determined according to preset mapping between the category and the score. According to the method, answer text semantic information can be effectively mined without depending on subjective question standard answers, and Chinese short text subjective question automatic scoring is achieved.

Description

technical field [0001] The present invention relates to the technical field of automatic review, specifically, it is a method and system for automatically scoring subjective questions of Chinese short texts using a long-short-term memory (LSTM, LongShort-Term Memory) neural network, which can be applied to answers given by Chinese natural language Automatic scoring of questions such as translation, short answer, judgment, and image-to-text conversion, and is finally applied to the review of homework and test papers and the learning evaluation process of students. Background technique [0002] Subjective questions occupy a very important position in subject learning and teaching. Its biggest advantage is that it can measure various complex behavioral goals, and it can also test students' creative thinking ability and expressive ability. Subjective questions have thus become one of the most widely used question types in subject teaching and testing. The heavy and mechanical s...

AI Technical Summary

Problems solved by technology

[0017] (1) The above method is mainly used for automatic scoring of English subjective questions, but due to the huge difference in natural language processing technology between Chinese and English, it is difficult to transplant the above method to automatic scoring of Chinese subjective questions

[0018] (2) The above method is aimed at closed questions, that is, there are standard answers to such questions, but in actual teaching and examination, many questions do not have standard answers

For such questions that have no standard answers and the scoring standards are relatively vague, the above algorithm is not applicable

[0019] (3) Most of the above methods rely on traditional language models, and the method of extracting text feature representation is complex, which cannot solve the problems of data sparsity and semantic sensitivity caused by short text length

Images