CN-122022082-A - Student comprehensive ability evaluation generation method and system based on artificial intelligence

CN122022082ACN 122022082 ACN122022082 ACN 122022082ACN-122022082-A

Abstract

The application relates to the technical field of intelligent teaching, in particular to a student comprehensive ability evaluation generation method and system based on artificial intelligence, wherein the method comprises the steps of acquiring multi-source learning process data and each ability dimension reference value in a specific learning period of a target student, and calculating each dimension ability deviation index; and judging whether the index exceeds a preset normal fluctuation interval, if not, integrating and analyzing multi-source data through a self-adaptive evaluation engine to generate a dynamic capacity evaluation report, if so, identifying key abnormal data items, generating a special stem pre-tuning strategy according to the key abnormal data items, and finally fusing the dynamic evaluation report and the intervention strategy to output a comprehensive capacity evaluation report containing development suggestions. The application is helpful for solving the problems of one-sided evaluation, lack of dynamic identification and closed-loop intervention in the prior art, and realizing intelligent, comprehensive and dynamic evaluation and accurate lifting support of the comprehensive capability of students.

Inventors

TAN WEIYONG
SUN JIAWANG
Li Manju

Assignees

湖南达美策略信息技术服务有限公司

Dates

Publication Date: 20260512
Application Date: 20260414

Claims (9)

1. The student comprehensive ability evaluation generation method based on artificial intelligence is characterized by comprising the following steps: acquiring multi-source learning process data of a target student in a specific learning period and each capacity dimension reference value corresponding to the multi-source learning process data; Calculating a capability deviation index of each capability dimension based on the multi-source learning process data and the corresponding capability dimension reference value; Judging whether the capability deviation index exceeds a preset normal fluctuation interval; If the capability deviation index does not exceed the normal fluctuation interval, carrying out integrated analysis on the multi-source learning process data based on a preset self-adaptive evaluation engine to generate a dynamic capability evaluation report of the target student; if the capability deviation index exceeds the normal fluctuation interval, identifying key abnormal data items from the multi-source learning process data; Generating a special dry pre-tuning strategy based on the key abnormal data item; and fusing the dynamic capacity evaluation report with a special dry pre-tuning strategy, and outputting a comprehensive capacity evaluation report containing development suggestions.
2. The method for generating the student comprehensive ability evaluation based on artificial intelligence according to claim 1, wherein generating the special stem pre-optimization strategy based on the key abnormal data item comprises: Acquiring numerical characteristics and associated dimensions based on the key abnormal data items; determining a development blocking grade corresponding to the key abnormal data item based on the numerical characteristics and the associated dimension; judging whether the development retardation grade reaches a preset intervention starting threshold value or not; if the development retardation level does not reach the intervention starting threshold, determining an adaptive adjustment upper limit and an adaptive adjustment lower limit of the associated dimension based on the capacity dimension reference value; invoking an arithmetic logic and a constraint framework based on the adaptive evaluation engine; combining the capability deviation index, the operation logic of a preset self-adaptive evaluation engine, the self-adaptive adjustment upper limit and the self-adaptive adjustment lower limit to generate a progressive tuning strategy as a special dry tuning strategy; if the development blocking level reaches the intervention starting threshold, generating a reconstruction type optimizing strategy based on the development blocking level and a constraint framework of a preset self-adaptive evaluation engine, and taking the reconstruction type optimizing strategy as a special stem optimizing strategy.
3. The method for generating an artificial intelligence based student comprehensive ability evaluation according to claim 2, wherein determining whether the level of development retardation reaches a preset intervention initiation threshold comprises: Identifying the core capacity short board type represented by the key abnormal data item; judging whether the core capacity short board type is a preset basic capacity short board type or not; If the core capacity short plate type is the basic capacity short plate type, judging that the development retardation grade reaches an intervention starting threshold; If the core capacity short board type is not the basic capacity short board type, counting the number of students with the same type of core capacity short board type and the total number of students in a specific learning period; calculating a population short board incidence rate based on the number of students and the total number of students for which the similar core capability short board types exist; judging whether the occurrence rate of the group short plates exceeds a first occurrence rate threshold value; And if the occurrence rate of the colony shortboards exceeds the first occurrence rate threshold, judging that the development retardation grade reaches the intervention starting threshold.
4. The method for generating the student comprehensive ability evaluation based on artificial intelligence according to claim 2, wherein generating the progressive optimization strategy by combining the ability deviation index, the operation logic of the preset adaptive evaluation engine, the adaptive adjustment upper limit and the adaptive adjustment lower limit comprises: judging whether the core capacity short board type is a preset movable capacity short board type or not; if the core capacity short board type is not the migratable capacity short board type, judging whether a comparable capacity development track formed by the target students in the history synchronization exists or not; If a comparable capacity development track exists, generating a progressive tuning strategy based on the time series characteristics of the comparable capacity development track and the key abnormal data item; if the comparability development track does not exist, generating a progressive tuning strategy based on the development retardation grade, the core capability short board type and the operation logic of a preset self-adaptive evaluation engine.
5. The method for generating the student comprehensive ability evaluation based on artificial intelligence according to claim 4, wherein the preset adaptive evaluation engine comprises a collaboration analysis module and an individual analysis module, and further comprises, after judging whether the core ability short board type is a preset migratable ability short board type: If the capability defect type is a potential constraint type, acquiring a learning situation mode of the target student; If the learning situation mode is a exploring cooperation mode, a gradual optimization strategy is generated by combining an adaptive adjustment upper limit, a capability deviation index and a cooperation analysis module of a preset adaptive evaluation engine; If the learning situation mode is an autonomous exercise mode, a progressive tuning strategy is generated by combining an adaptive adjustment lower limit, a capability deviation index and an individual analysis module of a preset adaptive evaluation engine.
6. The method of generating an artificial intelligence based student comprehensive ability assessment of claim 3, wherein determining that the level of progression block reaches the intervention initiation threshold if the occurrence of the crowd shortboards exceeds a first occurrence threshold comprises: If the occurrence rate of the group short plates exceeds a first occurrence rate threshold, judging whether the occurrence rate of the group short plates exceeds a systematic teaching risk threshold; If the occurrence rate of the group short plates does not exceed the systemic teaching risk threshold, judging that the development retardation grade reaches the intervention starting threshold; If the occurrence rate of the group short plates exceeds the systematic teaching risk threshold, adding teaching early warning marks for the core capacity short plate type and the corresponding student groups, and acquiring the accumulated triggering frequency of the teaching early warning marks; Judging whether the accumulated trigger frequency exceeds a preset frequency tolerance value or not; If the accumulated triggering frequency does not exceed the frequency tolerance value, judging that the development retardation level does not reach the intervention starting threshold value; if the accumulated triggering frequency exceeds the frequency tolerance value, judging that the development retardation grade reaches the intervention starting threshold value.
7. The method for generating the comprehensive student capacity evaluation based on artificial intelligence according to claim 6, wherein if the occurrence rate of the group short plates exceeds the systematic teaching risk threshold, adding teaching early warning marks for the core capacity short plate type and the corresponding student groups, and obtaining the accumulated trigger frequency of the teaching early warning marks comprises: if the occurrence rate of the group short plates exceeds the systematic teaching risk threshold, analyzing whether the associated learning task or the teaching link for generating the core capacity short plate type has consistency; If the associated learning task or the teaching link has consistency, adding a teaching early warning mark for the associated learning task or the teaching link, and updating the accumulated trigger frequency; if the related learning tasks or the teaching links do not have consistency, respectively calculating the short-board data duty ratio caused by different learning tasks or teaching links; determining dominant short-board cause duty cycle based on short-board data duty cycle; Judging whether the dominant short-board cause duty ratio exceeds a second duty ratio threshold value; If the dominant short plate cause ratio exceeds a second threshold ratio, adding a teaching early warning mark for a corresponding dominant learning task or teaching link, and updating the accumulated trigger frequency.
8. The method for generating the comprehensive ability evaluation of the student based on the artificial intelligence according to claim 1, wherein if the ability deviation index does not exceed the normal fluctuation interval, performing an integrated analysis on the multi-source learning process data based on a preset adaptive evaluation engine, and generating a dynamic ability evaluation report of the target student comprises: if the capability deviation index does not exceed the normal fluctuation interval, acquiring historical learning process data of a longer time span based on historical time sequence data contained in the multi-source learning process data; Analyzing the long-term development trend slope of each capability dimension based on the historical learning process data; judging whether the slope of the long-term development trend is lower than a preset development power early warning line or not; if the slope of the long-term development trend is lower than the development power early warning line, marking the capability dimension of the slope of the long-term development trend lower than the early warning line as the dimension to be excited, and judging whether a potential excitation resource library exists or not; If the potential excitation resource library exists, judging whether a matched potential excitation guiding scheme exists in the potential excitation resource library or not based on the dimension to be excited and the corresponding long-term development trend slope; If the matched potential excitation guiding scheme exists, generating a dynamic capability evaluation report based on the potential excitation guiding scheme; If the slope of the long-term development trend is not lower than the development power early warning line, carrying out integrated analysis on the multi-source learning process data of the current period based on a preset self-adaptive evaluation engine, and generating a dynamic capability evaluation report of the target student.
9. An artificial intelligence based student comprehensive ability evaluation generation system, comprising: The first acquisition module is used for acquiring multi-source learning process data of a target student in a specific learning period and each capacity dimension reference value corresponding to the multi-source learning process data; The computing module is used for computing the capability deviation index of each capability dimension based on the multi-source learning process data and the corresponding capability dimension reference value; the judging module is used for judging whether the capability deviation index exceeds a preset normal fluctuation interval; The report generation module is used for carrying out integrated analysis on the multi-source learning process data based on a preset self-adaptive evaluation engine to generate a dynamic capability evaluation report of a target student if the capability deviation index does not exceed a normal fluctuation interval; the abnormality identification module is used for identifying key abnormal data items from the multi-source learning process data if the capability deviation index exceeds a normal fluctuation interval; the strategy generation module is used for generating a special interference pre-tuning strategy based on the key abnormal data item; and the report output module is used for fusing the dynamic capacity evaluation report with a special stem pre-tuning strategy and outputting a comprehensive capacity evaluation report containing development suggestions.

Description

Student comprehensive ability evaluation generation method and system based on artificial intelligence Technical Field The application relates to the technical field of intelligent teaching, in particular to an artificial intelligence-based student comprehensive ability evaluation generation method and system. Background With the development of intelligent teaching and artificial intelligence technology, students' comprehensive ability evaluation has gradually changed from traditional single performance evaluation to procedural, multidimensional and personalized direction. At present, most teaching evaluation systems still use result data such as examination scores, homework accuracy and the like as core basis, and are insufficient in process data acquisition such as student classroom interaction, autonomous learning, collaborative exploration, practical operation and the like, so that real performances of multidimensional capabilities such as student knowledge mastering, logic thinking, autonomous learning, communication collaboration and the like are difficult to comprehensively reflect, and evaluation results are on the one side and labeled, and comprehensive capability development overall appearance of students cannot be accurately depicted. Meanwhile, the existing evaluation method mostly adopts a fixed evaluation model and a static evaluation standard, only can carry out result judgment on the capability level of students, lacks dynamic identification and abnormal analysis on capability expression fluctuation, and cannot distinguish whether the capability deviation of students belongs to normal learning fluctuation or abnormal capability short plates. When abnormal deviation occurs to a certain dimension of ability of students, the conventional system is difficult to automatically position key abnormal data and problem sources, a personalized dry pre-tuning strategy cannot be generated in a targeted manner, evaluation results and improvement guidance are disjointed, a closed loop of evaluation-analysis-intervention-lifting is difficult to form, and effective support cannot be provided for the ability development and teaching optimization of students. Disclosure of Invention In order to help solve the problems of one-sided and lack of dynamic identification and closed-loop intervention of the existing evaluation and realize intelligent, comprehensive and dynamic evaluation and accurate lifting support of the comprehensive capacity of students, the application provides an artificial intelligence-based student comprehensive capacity evaluation generation method and system. In a first aspect, the application provides an artificial intelligence based student comprehensive ability evaluation generation method, which adopts the following technical scheme: A student comprehensive ability evaluation generation method based on artificial intelligence comprises the following steps: acquiring multi-source learning process data of a target student in a specific learning period and each capacity dimension reference value corresponding to the multi-source learning process data; Calculating a capability deviation index of each capability dimension based on the multi-source learning process data and the corresponding capability dimension reference value; Judging whether the capability deviation index exceeds a preset normal fluctuation interval; If the capability deviation index does not exceed the normal fluctuation interval, carrying out integrated analysis on the multi-source learning process data based on a preset self-adaptive evaluation engine to generate a dynamic capability evaluation report of the target student; if the capability deviation index exceeds the normal fluctuation interval, identifying key abnormal data items from the multi-source learning process data; Generating a special dry pre-tuning strategy based on the key abnormal data item; and fusing the dynamic capacity evaluation report with a special dry pre-tuning strategy, and outputting a comprehensive capacity evaluation report containing development suggestions. Optionally, generating the special stem pre-tuning policy based on the key abnormal data item includes: Acquiring numerical characteristics and associated dimensions based on the key abnormal data items; determining a development blocking grade corresponding to the key abnormal data item based on the numerical characteristics and the associated dimension; judging whether the development retardation grade reaches a preset intervention starting threshold value or not; if the development retardation level does not reach the intervention starting threshold, determining an adaptive adjustment upper limit and an adaptive adjustment lower limit of the associated dimension based on the capacity dimension reference value; invoking an arithmetic logic and a constraint framework based on the adaptive evaluation engine; combining the capability deviation index, the operation logic of a preset self-