CN-121998489-A - Evaluation method and evaluation system of assessment object and electronic equipment

CN121998489ACN 121998489 ACN121998489 ACN 121998489ACN-121998489-A

Abstract

The application provides an evaluation method, an evaluation system and electronic equipment of an evaluation object, which relate to the technical field of communication and can be used for fusing multi-source data to analyze and evaluate and improving the accuracy and efficiency of an evaluation result of the evaluation object; the method comprises the steps of extracting target text information from multi-source data, wherein the target text information is used for reflecting the work performance of a target assessment object, generating an assessment prompt word of the target assessment object based on the target text information, target subjective assessment labels and an assessment knowledge base, wherein the assessment knowledge base is used for storing assessment rules and assessment cases corresponding to different subjective assessment labels, and obtaining an assessment result of the target assessment object by calling a large language model based on the assessment prompt word.

Inventors

Yang Hede
HE ZUQI
ZHANG YUNYUE
HU SHUYANG
YE YUNFENG
ZHANG JUN
Deng Junru

Assignees

南航数智科技(广东)有限公司

Dates

Publication Date: 20260508
Application Date: 20260115

Claims (10)

1. An evaluation method of an assessment object, the method comprising: acquiring multi-source data of a target examination object and a target subjective evaluation label; extracting target text information from the multi-source data, wherein the target text information is used for reflecting the work performance of the target assessment object; generating an evaluation prompt word of the target assessment object based on the target text information, the target subjective evaluation label and an evaluation knowledge base, wherein the evaluation knowledge base is used for storing evaluation rules and evaluation cases corresponding to different subjective evaluation labels; and based on the evaluation prompt word, obtaining an evaluation result of the target assessment object by calling a large language model.
2. The method of claim 1, wherein generating the evaluation prompt for the target assessment object based on the target text information, the target subjective evaluation label, and an evaluation knowledge base comprises: Retrieving an evaluation rule and an evaluation case corresponding to the target subjective evaluation label from the evaluation knowledge base; Integrating the target text information, the evaluation rule corresponding to the target subjective evaluation label and the evaluation case into a preset prompt word template to generate an evaluation prompt word of the target evaluation object.
3. The method of claim 1, wherein the extracting target text information from the multi-source data comprises: removing redundant data in the multi-source data to obtain initial data; converting the initial data into unified text data; Extracting the target text information from the text data through semantic analysis.
4. The method according to claim 1, wherein after obtaining the evaluation result of the target assessment object, the method further comprises: Pushing the evaluation result of the target examination object to an auditing terminal so that auditing personnel can audit the evaluation result of the target examination object; the auditing result is used for indicating whether the evaluation result of the target checking object passes the auditing or not; Under the condition that the auditing result indicates that the evaluation result of the target assessment object passes the auditing, determining that the evaluation result of the target assessment object is effective, and pushing the evaluation result of the target assessment object to a user terminal; And under the condition that the auditing result indicates that the evaluation result of the target assessment object fails the auditing, determining that the evaluation result of the target assessment object is invalid, and carrying out iterative correction on the evaluation result of the target assessment object until the evaluation result of the target assessment object passes the auditing.
5. The method of claim 4, wherein iteratively modifying the evaluation of the target assessment object comprises: In the process of each iteration, correcting the evaluation prompt word based on the correction opinion of the assessment personnel to obtain a corrected evaluation prompt word; and obtaining a corrected result of the evaluation result of the target assessment object based on the corrected evaluation prompt word.
6. The method according to claim 4, wherein the method further comprises: The method comprises the steps of adjusting the auditing priority of the evaluation result of a target evaluation object based on the confidence of the evaluation result of the target evaluation object, wherein the confidence is determined based on the semantic matching degree between target text information corresponding to an evaluation prompt word of the target evaluation object and the target subjective evaluation label, and the semantic matching degree is determined based on the large language model.
7. The method according to claim 1, wherein the method further comprises: And under the condition that the updating operation of the evaluation knowledge base is detected, generating a new prompt word template based on the updating content of the evaluation knowledge base, wherein the updating content comprises a new subjective evaluation label, and an evaluation rule and an evaluation case corresponding to the new subjective evaluation label.
8. The evaluation system of the assessment object is characterized by comprising a multi-source data processing module, an intelligent body and a large language model calling module; The multi-source data processing module is used for acquiring multi-source data of a target assessment object, extracting target text information from the multi-source data and sending the target text information to the intelligent agent; the intelligent agent is used for acquiring a target subjective evaluation label and the target text information of the target evaluation object and generating an evaluation prompt word of the target evaluation object based on the target text information, the target subjective evaluation label and an evaluation knowledge base; The large language model calling module is used for calling a large language model, inputting the evaluation prompt word of the target assessment object into the large language model to obtain the evaluation result of the target assessment object output by the large language model, and returning the evaluation result of the target assessment object to the intelligent agent.
9. The system for evaluating an assessment object according to claim 8, wherein the system further comprises an evaluation knowledge base module, a result auditing module, and a result output module; the evaluation knowledge base module is used for storing evaluation rules and evaluation cases corresponding to different subjective evaluation labels; The result auditing module is used for auditing the evaluation result of the target examination object; the result output module is used for pushing the evaluation result of the assessment object to the user terminal.
10. An electronic device, comprising a processor and a memory; the memory stores instructions executable by the processor; The processor is configured to, when executing the instructions, cause the electronic device to implement the method of any one of claims 1-7.

Description

Evaluation method and evaluation system of assessment object and electronic equipment Technical Field The present application relates to the field of communications technologies, and in particular, to an evaluation method, an evaluation system, and an electronic device for an assessment object. Background Along with the continuous promotion of enterprise digital transformation, the work assessment field of management staff is gradually updated from a traditional manual experience driving mode to a data driving mode, and subjective assessment of staff is used as a core carrier for describing unquantized characteristics of an assessment object, so that the assessment quality and efficiency of the assessment object directly influence the establishment of accurate portrait construction and efficient screening work of the assessment object. However, the existing evaluation mode still has a plurality of defects, and the requirements of personnel checking work on high precision, high efficiency and high consistency of the labels are difficult to meet. Disclosure of Invention The invention aims to provide an evaluation method, an evaluation system and electronic equipment of an evaluation object, which can integrate multi-source data to analyze and evaluate, improve the accuracy and efficiency of the evaluation result of the evaluation object, The application provides an evaluation method of an evaluation object, which is characterized by comprising the steps of obtaining multi-source data and target subjective evaluation labels of the target evaluation object, extracting target text information from the multi-source data, enabling the target text information to reflect work performance of the target evaluation object, generating evaluation prompt words of the target evaluation object based on the target text information, the target subjective evaluation labels and an evaluation knowledge base, wherein the evaluation knowledge base is used for storing evaluation rules and evaluation cases corresponding to different subjective evaluation labels, and obtaining an evaluation result of the target evaluation object by calling a large language model based on the evaluation prompt words. According to the evaluation method for the evaluation object, provided by the embodiment of the application, the multi-source data and the target subjective evaluation label are combined, the target text information reflecting the working performance of the evaluation object is extracted, the evaluation prompt word is generated by using the evaluation knowledge base storing the evaluation rules and cases, and then the large language model is called to output the evaluation result, so that the comprehensiveness and the accuracy of the evaluation basis can be improved by combining the multi-dimension data, the consistency and the rationality of the evaluation logic can be guided and ensured by the standardization of the evaluation knowledge base, and the overall efficiency of the evaluation work can be effectively improved by means of the automatic generation capability of the large language model. In some embodiments, based on target text information, target subjective evaluation labels and an evaluation knowledge base, an evaluation prompt word of a target assessment object is generated, and the method comprises the steps of retrieving evaluation rules and evaluation cases corresponding to the target subjective evaluation labels from the evaluation knowledge base, integrating the target text information, the evaluation rules and the evaluation cases corresponding to the target subjective evaluation labels into a preset prompt word template, and generating the evaluation prompt word of the target assessment object. In some embodiments, extracting the target text information from the multi-source data includes removing redundant data from the multi-source data to obtain initial data, converting the initial data into unified text data, and extracting the target text information from the text data by semantic analysis. In some embodiments, after the evaluation result of the target assessment object is obtained, the method further comprises pushing the evaluation result of the target assessment object to an auditing terminal so that an auditing person can audit the evaluation result of the target assessment object, receiving an auditing result fed back by the auditing terminal, wherein the auditing result is used for indicating whether the evaluation result of the target assessment object passes the auditing, determining that the evaluation result of the target assessment object is effective when the auditing result indicates that the evaluation result of the target assessment object passes the auditing, pushing the evaluation result of the target assessment object to a user terminal, determining that the evaluation result of the target assessment object is invalid when the auditing result indicates that the evaluation result of the target ass