Predicting Consumer Personalities from What They Say

This study mapped personality based on the newly proposed extraction method from consumers’ textual data and revealed relevance (attention) polarity (affection) of words associated with a specific trait. Furthermore, we illustrate how unique are used to predict consumer’s behavior certain traits. In this study, employed scales Kaggle MBTI Personality dataset examine methodology’s effectiveness, extract traits into features, map them traits/dimensions existing scale. Based results obtained in assert that using TF-IDF algorithm is good way generate custom dictionary. sentiment scoring an AI-empowered machine learning provides useful filter validate more coherent understand and, thus, communicate particular aspect personality. Finally, four situations involving interaction between attention (frequency) affection (sentiment) allow us better consumer use feature terms (TF-IDF score) (sentiment score).