TensorFlow-Based Automatic Personality Recognition Used in Asynchronous Video Interviews


With the development of artificial intelligence (AI), the automatic analysis of video interviews to recognize individual personality traits has become an active area of research and has applications in personality computing, human-computer interaction, and psychological assessment. Advances in computer vision and pattern recognition based on deep learning (DL) techniques have led to the establishment of convolutional neural network models that can successfully recognize human nonverbal cues and attribute their personality traits with the use of a camera. In this paper, an end-to-end AI interviewing system was developed using asynchronous video interview (AVI) processing and a TensorFlow AI engine to perform automatic personality recognition (APR) based on the features extracted from the AVIs and the true personality scores from the facial expressions and self-reported questionnaires of 120 real job applicants. The experimental results show that our AI-based interview agent can successfully recognize the “big five” traits of an interviewee at an accuracy between 90.9% and 97.4%. Our experiment also indicates that although the machine learning was conducted without large-scale data, the semisupervised DL approach performed surprisingly well with regard to APR despite the lack of labor-intensive manual annotation and labeling. The AI-based interview agent can supplement or replace existing self-reported personality assessment methods that job applicants may distort to achieve socially desirable effects.


• Resume classification: resume will be analysed with various machine learning algorithm like naïve bayes,svm or random forest. Effectiveness with the oraganization will be analysed. • Tone analysis: User voice tone will be classified by converting it into text . and text analyzation will identify tone of user with various machine learning algorithm like naïve bayes,svm or random forest. Which ultimately will identify effectiveness for our organization . • Video based face feature analysis: With the help of CNN algorithm and face landmark user face will be captured by camera and feature extraction of face will be done which will result in obtaining output such as happy face,entertaining face,good gesture,good smile.Then effectiveness with the job location will be identified. • CNN model generation: face categories will be added and trained with convolution neural network. • Text model generation:Random forest ,naïve bayes ,svm model will be prepared for tone analysis. • Face recognition:- for employee verification we can use face recognition module associated with azure microsoft. • User Registration: User has to register to check the interview performance . • User Login: User can login to system to check the interview performance. • Admin: Admin adds the user and their details.


CNN,Random forest ,naïve bayes ,svm

Software And Hardware

Software Requirements: • Windows 10, Windows 7 • Python,anaconda,jupyter notebook,spyder • Back end:-mysql,sqlyog • tensorflow Hardware Components: • Processor – i3,i5 • Hard Disk – 5 GB • Memory – 6GB RAM