Machine intelligent reviewing method and system for simple answers

Abstract

The invention relates to test paper reviewing, and discloses a simple answer machine intelligent reviewing method and system, which can avoid misjudgment of behaviors of obtaining high scores throughword stacking, so that the scoring result is more objective and accurate. The method comprises the steps: acquiring subject keywords and common keywords of a subject in advance based on subject corpora, acquiring an associated word set of the subject keywords, and constructing a keyword library of the subject; obtaining answering information and standard answers of the target test questions; basedon a keyword library, extracting a subject keyword set and a common keyword set in the standard answer, and determining an associated word set of each subject keyword to expand the subject keyword set; identifying subject keywords and associated words in the answering information based on the expanded subject keyword set, and identifying common keywords in the answering information based on the common keyword set; calculating sentence reasonability of the answering information; and calculating the score of the answering information according to a scoring formula.

Description

technical field [0001] This application relates to test paper review, in particular to machine intelligence review technology for short-answer questions. Background technique [0002] Examination is an indispensable and important part of teaching activities. It is an inspection of students' usual learning situation and a test of teachers' teaching level. [0003] With the development of computer technology, computer-aided examination papers are generally used to evaluate test papers. At present, the automatic evaluation technology of objective questions is relatively mature, while the automatic evaluation of subjective questions (such as short answer questions) is still a difficult point in computer automatic evaluation technology. Although some automatic evaluation methods for subjective questions have been proposed, there are still many problems in these methods. For example, the Chinese application whose publication number is CN108959261A discloses a device and method fo...

AI Technical Summary

Problems solved by technology

The problem that this device exists is: calculate the word similarity of keyword only by word meaning similarity, and adopt the word similarity of the words contained in two sentences to calculate the semantic similarity of two sentences, only the appearance information of sentence lack of judgment on the logic of the sentence itself, it is easy to cause misjudgment such as obtaining high scores by piling up words

Images