Personality evaluation method and system based on emotional states and emotional changes

Abstract

The invention discloses a personality evaluation method and system based on emotional states and emotional changes, and the method comprises the steps: collecting a video of a testee during the current speech or report; preprocessing the video to obtain preprocessed video data, voice data and semantic data; calling trained emotion recognition model E-R-Model to test the preprocessed video data, voice data and semantic data to obtain an emotion type of the testee and a distribution result along time; calling personality recognition model P-R-Model to analyze the emotion type of the testee and the distribution result along the time to obtain a personality result of the testee. In the embodiment of the invention, the personality type of the testee is evaluated according to the emotional stateand the emotional change of the testee in the speech or report process, so that the test time can be saved, and the test objectivity is improved.

Description

technical field [0001] The present invention relates to the technical fields of machine learning machines and deep learning, in particular to a personality assessment method and system based on emotional states and emotional changes. Background technique [0002] Under the background of new industrialization and informationization, personalized education is one of the effective ways to improve the quality of education. One of the necessary conditions for the effective implementation of personalized education is that each student must "know himself" deeply and precisely, that is, know himself. After the popularization of higher education, this is precisely the problem of education and even the whole society in the new era. Students themselves and teachers cannot understand each student, so that personalized education cannot be achieved. According to each person's different personality, implementing different education types and education methods for each person is a necessa...

AI Technical Summary

Problems solved by technology

[0005] The above-mentioned technology has the following disadvantages: (1) Questionnaire: the first, the accuracy of the test, a large part is determined by the design of the questionnaire; Fill in randomly or intentionally by your own ideas

Questionnaires have the disadvantages of low test authenticity, small test scope, and lack of flexibility

(2) Observation category: The professional level of the observer determines the accuracy of the test, and the preference of the observer will lead to different observation results; the observation category needs to observe each tester, so it will consume a lot of manpower and time

The observation category has the disadvantages of consuming a lot of time, a lot of manpower, unstable test results, and difficult to promote; (3) machine category: only from paper questions to filled in by webpages and APPs, the essence is that the testees do the questions Or play a game to evaluate the subject's personality based on the results, so the accuracy rate is not high

The current machine class has shortcomings such as low accuracy and consuming a lot of tester time

Images