The temperament of individuals could also be probably analyzed mistreatment the content they post, mistreatment profile image selection and their social media sites. Throughout this paper, we have a tendency to investigate Twitter profiles pictures to predict temperament of three classes of the general public. For the analysis, we got distributed the individuals into three major classes (like Business Bodies, General public and Indian geneticist). Each square measure collected the Twitter profile knowledge and keeps them in MongoDB knowledge base. This assortment of information sets is any analyzed with their profile pictures mistreatment options of color, image type, image composition and their on-line behaviour. Each square measure predicting regarding their temperament traits mistreatment their posts and on-line engagement on Twitter. For the interpretability, our analysis targeted aesthetic and facial expression to explore temperament traits. Our results predict the behaviour of Twitter users supported the profile image they choose. As an example, conscientious and agreeable users show positive emotions, whereas users having openness choose aesthetic photos. Performance was achieved by D-CNN, wherever the accuracy measures regarding 98 to 99%. Sensation Neural Network (SNN) technique was tried that we got achieved 99.157% correct prediction and encounter that temperament prediction is tougher for strong accuracy.
Volume 11 | 07-Special Issue
Pages: 278-285