CN-121435959-B - Text evaluation method and device, electronic equipment and storage medium

CN121435959BCN 121435959 BCN121435959 BCN 121435959BCN-121435959-B

Abstract

The application discloses a text evaluation method, a text evaluation device, electronic equipment and a storage medium. The method comprises the steps of obtaining a text to be evaluated, determining all reference fields in the text to be evaluated and document identification numbers corresponding to the reference fields according to reference marks in the text to be evaluated, determining at least one text attribute and at least one reference attribute of the text to be evaluated according to the reference fields and the corresponding document identification numbers, and evaluating the text to be evaluated according to the text attributes and the reference attributes. According to the technical scheme provided by the embodiment of the application, different text attributes and different quotation attributes can represent the quality of the text to be evaluated from different dimensions, comprehensively evaluate the text to be evaluated, and can improve the accuracy of text evaluation.

Inventors

JIANG JIARUI
QUAN SHU
Ye Sihui

Assignees

解螺旋(上海)科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260105

Claims (7)

1. A text evaluation method, comprising: acquiring a text to be evaluated generated by a large language model; determining all the reference fields in the text to be evaluated and the document identification numbers corresponding to the reference fields according to the reference marks in the text to be evaluated; Determining at least one text attribute and at least one reference attribute of the text to be evaluated according to each reference field and the corresponding document identification number, wherein the reference attribute comprises the number of reference documents and the number of reference times, and the number of reference times is the number of the reference fields; Evaluating the text to be evaluated according to the text attributes and the reference attributes; Wherein, the quotation attribute also comprises quotation correct rate and quotation correct rate; Determining at least one referencing attribute of the text to be evaluated according to each referencing field and the corresponding document identification number, including: searching in a preset document database according to each document identification number to determine a correct document reference item in each reference field; determining content correct reference items according to each document correct reference item and the corresponding reference document; Determining the quotation accuracy according to each content correct quotation term and the quotation document number; determining the reference accuracy according to each content correct reference item and the reference times; The searching in a preset document database according to each document identification number to determine the correct reference item of the document in each reference field comprises the following steps: For any document identification number, searching in the document database according to the document identification number, and judging a reference field corresponding to the document identification number as the correct reference item of the document if document metadata corresponding to the document identification number exists in the document database; If the document identification number does not exist in the document database or document metadata corresponding to the document identification number does not exist, judging a reference field corresponding to the document identification number as a document fictive reference item; The reference attribute further includes an imaginary document rate; Determining at least one referencing attribute of the text to be evaluated according to each referencing field and the corresponding document identification number, including: And determining the fictional literature rate according to the fictional literature reference item and the number of the cited literature.
2. The method of claim 1, wherein said determining at least one referencing attribute of said text under evaluation based on each of said referencing fields and corresponding said document identification number comprises: determining the reference times in the text to be evaluated according to each reference field; and checking the weight according to the document identification number, and determining the number of the cited documents in the text to be evaluated.
3. The method of claim 1, wherein said determining content correct reference based on each of said document correct reference and corresponding reference comprises: based on a preset semantic discrimination model, comparing the correct reference item of the document with the semantics of the corresponding reference document, and if the comparison result is the same as the semantics, judging the correct reference item of the document as the correct reference item of the content.
4. The method of claim 1, wherein said determining text attributes of said text to be evaluated based on said text to be evaluated and each of said reference fields comprises: determining the number of text words and the number of reference words of the text to be evaluated according to the text to be evaluated and each reference field; And taking the number of text words and the number of reference words as the text attribute.
5. A text evaluation apparatus, comprising: the text acquisition module is used for acquiring a text to be evaluated generated by the large language model; The field determining module is used for determining all the reference fields in the text to be evaluated and the document identification numbers corresponding to the reference fields according to the reference marks in the text to be evaluated; The attribute determining module is used for determining at least one text attribute and at least one reference attribute of the text to be evaluated according to each reference field and the corresponding document identification number, wherein the reference attribute comprises the number of reference documents and the number of reference times, and the number of reference times is the number of the reference fields; The attribute evaluation module is used for evaluating the text to be evaluated according to the text attributes and the reference attributes; wherein, the quotation attribute also comprises quotation correct rate and quotation correct rate; the attribute determination module includes: A document correct reference item determining unit, configured to perform searching in a preset document database according to each document identification number, and determine a document correct reference item in each reference field; A content correct reference item determining unit configured to determine a content correct reference item based on each of the document correct reference items and the corresponding reference document; The quotation correct rate determining unit is used for determining the quotation correct rate according to each content correct quotation item and the quotation document number; a reference accuracy determining unit, configured to determine a reference accuracy according to each content correct reference item and the number of references; the document proper reference item determination unit includes: A correct reference item determination subunit configured to search, for any one of the document identification numbers, in the document database according to the document identification number, and if document metadata corresponding to the document identification number exists in the document database, determine a reference field corresponding to the document identification number as the correct reference item for the document; A fictitious reference item determination subunit configured to determine, if the document identification number does not exist in the document database or there is no document metadata corresponding to the document identification number, a reference field corresponding to the document identification number as a fictitious reference item; wherein the reference attributes further include an imaginary document rate; the attribute determination module includes: and an imaginary document rate determining unit configured to determine the imaginary document rate based on the document imaginary reference item and the number of cited documents.
6. An electronic device, the electronic device comprising: at least one processor, and A memory communicatively coupled to the at least one processor, wherein, The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the text evaluation method of any of claims 1-4.
7. A computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, the computer instructions for causing a processor to perform the text evaluation method of any of claims 1-4 when executed.

Description

Text evaluation method and device, electronic equipment and storage medium Technical Field The present application relates to the field of data processing technologies, and in particular, to a text evaluation method, a text evaluation device, an electronic device, and a storage medium. Background With the continuous development of society and the long progress of technology, various fields of various industries have developed more and more academic texts such as academic articles or news manuscripts. These academic texts are of varying quality and can be evaluated for ease of reading, learning, and discussion by those skilled in the relevant arts. Therefore, how to evaluate academic text is one of the important points of research by those skilled in the relevant arts. Currently, few means are used for checking and evaluating the cited references in academic texts, and the auditors usually review and judge the references manually, so that the efficiency is poor, and the evaluation accuracy is also problematic. Disclosure of Invention The application provides a text evaluation method, a text evaluation device, electronic equipment and a storage medium, which are used for improving the accuracy of text content evaluation. According to an aspect of the present application, there is provided a text evaluation method including: Acquiring a text to be evaluated; determining all the reference fields in the text to be evaluated and the document identification numbers corresponding to the reference fields according to the reference marks in the text to be evaluated; determining at least one text attribute and at least one reference attribute of the text to be evaluated according to each reference field and the corresponding document identification number; and evaluating the text to be evaluated according to each text attribute and each reference attribute. According to another aspect of the present application, there is provided a text evaluation apparatus including: the text acquisition module is used for acquiring a text to be evaluated; The field determining module is used for determining all the reference fields in the text to be evaluated and the document identification numbers corresponding to the reference fields according to the reference marks in the text to be evaluated; the attribute determining module is used for determining at least one text attribute and at least one quoting attribute of the text to be evaluated according to each quoting field and the corresponding document identification number; and the attribute evaluation module is used for evaluating the text to be evaluated according to the text attributes and the reference attributes. According to another aspect of the present application, there is provided an electronic apparatus including: at least one processor, and A memory communicatively coupled to the at least one processor, wherein, The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the text evaluation method of any one of the embodiments of the present application. According to another aspect of the present application, there is provided a computer readable storage medium storing computer instructions for causing a processor to execute a text evaluation method according to any one of the embodiments of the present application. According to another aspect of the application, a computer program product is provided, comprising a computer program which, when executed by a processor, implements a text evaluation method according to any of the embodiments of the application. According to the technical scheme, the method and the device for evaluating the text of the large language model, the reference marks in the text to be evaluated are generated for identification, all the reference fields and the document identification numbers corresponding to the reference fields in the text to be evaluated are determined, the positions and the specific contents of the reference fields in the text to be evaluated can be rapidly located, the corresponding document identification numbers are rapidly searched for determining the reference attributes, at least one text attribute and at least one reference attribute of the text to be evaluated are determined according to the reference fields and the corresponding document identification numbers, the text attribute and the reference attribute are obtained through analysis of the text to be evaluated, basis is provided for evaluating the text, different text attributes and different reference attributes can represent the quality of the text to be evaluated from different dimensions, the text to be evaluated is comprehensively evaluated, and the accuracy of text evaluation can be improved. It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the application or to delineate the scope of the application. Other