An Analysis Of Students' Personality Traits Based On Face Features And Deep Learning

With the high-speed development of artificial intelligence,sped up the process of global economic integration,competitions have passed through boundaries,and it has a great impact on China's economy and society.Global competition is essentially the competition of human resources.Consequently,the social demand for high-quality per sons of ability are growing.As important exporters of persons of ability,Colleges and universities are playing increasingly important roles.How to cultivate a high-quality e mployment team in the background of all kinds of pressure and contradictions also be comes an urgent problem for colleges and universities to be solved.The Big Five personality traits and facial feature analysis based on Deep Learnin g aims to use psychology,human resource management and other professional backgr ounds,introducing the method of personality tests,and the technology of face recogni tion and machine learning,to find the relationships between the Big Five personality t raits and facial feature,which can be used as auxiliary tools for education managers to identify the five personality traits preliminaries,and providing further reference for st udents' personality traits analysis according to the Big Five personality traits.The mai n work we did to solve the above problems,and achieve research objectives is as follo ws:According to the theory of psychological personality test,we think that human personality traits are related to some characteristics of human face in some way.We introduced the NEO-PI-R,to test on the big five personality traits by using statistical analysis method,and then propose an analysis study of the Big Five Personality traits based on face feature and Deep Learning.The Big Five Personality scale can provide relatively real and reliable personality traits,which lays the foundation for the accuracy of the experiment.In order to correlate the personality traits of the subjects with the facial features,we extracted the facial feature for each of the subjects.We introduced the improved ASM,which has no need to consider the advantages caused by the size of images,to extract the facial feature points,and select 32 representative features as facial feature data,which including the ratio of the distance between the features to the entire faceratio and the aspect ratio of the facial features,so that we can provide training bases for Deep Learning.We use the method of Deep Learning to practice and classify the students' personality and facial features through the model of mutual understanding in network,so as to find the relationship between the five personality traits and the facial features.Due to Deep Learning with unsupervised learning classification properties,the method based on face features and Deep Learning provided in this paper uses training samples to update the relevant weights.Compared with the traditional method of personality test,the test results show that this algorithm has a better efficacy in terms of timeliness and accuracy.