A Hybrid Method of N-Gram and Bag-of-Words (BoW) Models on the Assessment of Adolescent Personality Traits Through the Application of the Naïve Bayes Algorithm
Irzal Arief Wisky, Sarjon Defit, Gunadi Widi Nurcahyo · 2024
The integration of information technology into scientific disciplines, including Psychology, has been transformative, particularly with the advent of artificial intelligence in the form of Expert Systems. This research focuses on utilizing Natural Language Processing (NLP) techniques and classification methods to accurately delineate adolescent personality types. By employing advanced methodologies such as the Bag-of-Words (BoW) model, N-gram configurations, and preprocessing techniques like Rapid Automatic Keyword Extraction (RAKE), this study aims to improve the precision of personality classification in Indonesian adolescents. Our rigorous experiments reveal a significant enhancement in classification accuracy using the BoW approach, coupled with unigram, bigram, and trigram configurations, compared to traditional methods. This empirical evidence underscores the potential of these innovative techniques in providing more precise insights into adolescent personality archetypes. The novelty of this study lies in its application of combined NLP strategies within the Indonesian context, which has yielded substantial improvements in classification performance, highlighting the efficacy and promise of these approaches in psychological assessments.