Mining X Platform Data for Predictive Personality Modeling رسالة ماجستير

Abstract

Social media has gained widespread acceptance and is now a prominent feature of people’s modern digital landscape. A significant portion of human endeavors, including interactions with others, entertainment consumption, online shopping, and information seeking, are increasingly conducted through digital platforms and devices. Previous studies have revealed that more than 2.4 billion people of all ages use social media, and they spend a significant amount of their time engaged in it. Engaging in actions such as updating profiles, posting status updates, and sharing various content offers users a means to express a substantial amount about themselves. They showcase their personalities by using self-descriptions, sharing their current statuses, uploading pictures, and highlighting their interests. Social media profiles and tweets frequently serve as channels through which users provide glimpses into their traits and characteristics. This study addresses the predictive capacity of social media profiles regarding personality traits. This research work analyzed the data generated from interactions on social media platforms, intending to uncover connections and associations between online actions and diverse personality traits. Data analysis methodologies such as natural language processing and deep learning methods are utilized to examine elements such as language utilization, content inclinations, and interaction tendencies. Through these techniques, the study seeks to understand potential indicators of personality traits. In this study, utilizing deep learning algorithms such as Long Short-Term Memory (LSTM) and Gated Recurrent Units (GRUs) models after cleaning a dataset of X Platform posts. We found that these tools can be utilized to specify users' personalities by analyzing the Tweets users' posts. A 90% accuracy of predicting user personality was achieved. Consequently, this VI project is evidence of the possibility of utilizing social media data mining to reach more accurate results and predict the changes in the users' personalities.