Analyzing Job Interview Performance Prediction using Graph Features and Convolutional Neural Networks

The psychological reactions of candidates to personnel selection are becoming more important since they significantly impact the outcome of the selection process. In this study, we created and evaluated the Multi-dimensional Job Interview Self-Efficacy Scale (MJISE). Five components of interviewees' self-efficacy form the basis of our proposed MJISE: self-promotion, dealing with difficult interviewers, interview-related anxiety, and content and practical interview preparation.This proposed approach consists of three stages, which are preprocessing data, training the model, and extracting features. When it comes to text classification, preprocessing is king. In order to make text data useable by a classifier, it is necessary to extract a set of words from the text data and convert them into a feature set. Throughout the training process, it employed a GCNN. With a 91.30% average accuracy, our innovative approach surpasses CNN and ANN.