Text modeling for Prediction of Person’s Big-5 Traits using 2-LSTM Deep Neural Network

Abstract

Personality recognition models are useful in different domains, such as learning via electronic media, information retrieval, relationship, and electronic commerce through the user interface that follows the participation of user personality. Summarizing last user communications is just a beginning stage in describing consumer behavior from a personality perspective. The "Big Five" model of personality measurements have appeared in current years as particular very much examined proportions of personality formation. A person thought is expressed by what he writes about any condition. Hereafter personality traits prediction is the important area of research. This area of research comes under the NLP (Natural Language Processing). In this research we proposed to utilize the time effective sentence tokenization algorithm, effective text pre-processing prominence on emoji’s, LSTM for null text generation, followed by Two LSTM layers based deep neural network learning classifier. Eventually we have achieved significant accuracy as 64%.