Enhancing Personality Classification through Textual Analysis: A Deep Learning Approach Utilizing MBTI and Social Media Data

• Published in: 2023 International Conference on Network, Multimedia and Information Technology (NMITCON) pp. 01 - 06
• Main Authors: Mohan, G Bharathi, Kumar, R Prasanna, R, Elakkiya, Gorantla, Snehitha
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01-09-2023
• Subjects: Computational modeling; Data models; Deep learning; Human-robot interaction; Information technology; LSTM; MBTI; Recommender systems; Social networking (online); word embedding; Word2Vec
• DOI: 10.1109/NMITCON58196.2023.10276193

Personality computing has grown significantly in use, with useful applications emerging in fields of study like human-robot interaction and recommendation systems. Traditional recommendation systems frequently run into issues with free riders, data scarcity, and a lack of understanding of user preferences. By strengthening the grasp of user preferences and increasing the precision of recommendations, the addition of well-established user personality features aids in resolving these problems. Therefore, the goal of this research is to take advantage of personality computing's numerous advantages while overcoming the drawbacks of current recommendation systems. This model achieved an accuracy of 97% on the test data because it was trained on pre-processed and padded sequences. By utilising LSTM layers, the model is better able to understand contextual information included in the text input and efficiently capture sequential dependencies.