Adaptive evaluation method and evaluation system for cognitive competence and use method of evaluation system

Abstract

The invention provides an adaptive evaluation method for cognitive competence, which is called as a hierarchical adaptive optimization project reaction theoretical model, and the method comprises thesteps of if the tested competence is at the beginning of a test, endowing the initial prior distribution of the tested competence with an initial value by a Bayesian hierarchical model; if the n-1 tested capacities are tested, and when the nth tested capacity is subjected to (t-1) times of testing and needs to be subjected to the tth time of testing, assigning the tested capacity prior distribution by posteriori distribution obtained after (t-1) times of testing; solving a candidate design for maximizing the mutual information utility of the expected utility function, and testing the tested capability by using test questions corresponding to the candidate design; calculating the distribution value of each unknown parameter in the IRT model; obtaining an observation result; calculating posterior distribution of the tested capability; judging whether the current total information amount meets a termination condition, and if the current information amount is greater than the threshold condition, ending the test to obtain an observation result; and if the current information amount does not meet the termination condition, taking the posteriori distribution of the tested characteristicsas the prior distribution of the tested characteristics in the next question test, and continuing circulation.

Description

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine

Login to view more

AI Technical Summary

Problems solved by technology

But its disadvantages are: (1) Due to the repeated selection of the topic with the largest information gain, most of the topics in the question bank cannot be selected, and there is a risk of leaking the question bank

(2) When the prior distribution of the subject's ability is not clear or the answering data at the beginning is too small to determine the true ability of the subject, rapid convergence will cause the measurement results to deviate from the true value

(3) There are no other auxiliary parameters to help correct the fitting direction, which may easily cause the measurement results to deviate from the true value

[0054] (1) The question-making strategy cannot be adjusted according to the ability of the respondent, and the evaluation result must be obtained after answering all the questions at one time. The measurement time is too long and the efficiency is low

[0055] (2) It cannot reflec

Images