A Classification Pipeline Methodology for Predicting Personality Traits

With multiple advancements in machine learning and deep learning a multitude of tasks that were either too complex to hardcode or even regarded as impossible can now be performed. This technology can be used to our advantage to determine someone's personality based solely on their perception and opinions. Nowadays, a lot of tasks involving the victim's personality are necessary for criminal activity surveillance, personality has started to play a significant role in mental health therapy as well as hospitality businesses are trying to monitor the personality of customers to provide better services. We can assist in the completion of such a work by combining deep learning with a classification-based machine learning model connected by a data pipeline. The use case can also be extended to hospitality businesses to understand the mindset of the patient and is also useful during selection of candidates for a particular job profile based on their personality traits. It is also used in many successful marketing campaigns to target specific audiences. We propose a combined classification pipeline that employs both unsupervised and supervised deep learning approaches such as K-Means along with LSTM, GRU, RNN and Autoencoders (Deep Embedded Clustering) to draw a comparative analysis between these approaches to analyze personalities successfully.