Human Personality Analysis and Enhancer using AI

This research explores the application of logistic regression for classifying individuals into five distinct personality types: responsible, serious, extraverted and lively and dependable. The classification is based on a dataset comprising 709 individuals, each of whom is evaluated on five key personality traits: PEN affiliative traits include: openness to experience, negative emotionality, conscientiousness, agreeableness, and extraversion. These traits are known in psychology as the important factors that define behaviour and predispositions of the persons. The logistic regression analysis, often used and easy to explain, is applied to the analysis of the links between these personality traits and the respective personality type. The model aims at identifying personality type given certain interaction and level of importance of the traits in question. An important advantage of using logistic regression is the provision of probability values for each class, with results in class predictions as well as their likelihood. For its assessment, the model uses more than one key performance indicator: accuracy, precision, recall, and F1score which provide an objective picture of the effectiveness of the model. The proposed logistic regression classifier produces an overall accuracy of 76 percent with reasonable precision and recall rates for each of the five personality classes. These outcomes lead to the conclusion that logistic regression can be considered as an effective and accurate instrument of personality classification. In addition, the model comes with interpretability, the ability to explain the outcomes of each trait with regards to the probability of a certain personality classification.