A Psycholinguistic Approach to Career Selection Using NLP with Deep Neural Network Classifiers

Abstract

Career direction is a crucial matter not to be undermined in the development of a more efficient generation of the corporate workforce. In order to obtain accurate career direction, one would think of different ways of identifying attributes that would lead to an accurate classification of personality. In this paper, the goal is extracting personality from the use of language. The paper covers all aspects of this process in terms of Text Normalization Techniques, Feature Extraction, Feature Selection, Data Pre-Processing, Data Sampling, Training Predictive Models to predict personality types, validating the results on test data, and finally, and finally,compare the findings with other approaches to personality classification. After having a personality type classified, the process is as simple as matching career paths that are most likely suitable for the user. All these processes combined by experimenting with various approaches to each operation would result in personality attribute classifiers yielding an average of 96% accuracy.