Author(s): O Ramya Teja, Chikoti Likitha, Challa Veda Sri and Fathima Anjum

Abstract: The comprehension of novelty seeking (NS), an innate personality feature that affects travel motivation and location selection, requires an awareness of the experience related information found in online travel evaluations. Because of their large number and lack of organization, manually classifying these evaluations is difficult. The objective of this research is to address these shortcomings by creating a deep learning model and classification system. Four dimensions—relaxation seeking, experience seeking, arousal seeking, and boredom alleviation—that were combined from earlier research were included in a multi-dimensional categorization framework for the NS personality characteristic. Using 30 000 TripAdvisor reviews as a basis, we suggest a deep learning model that uses the Bidirectional Encoder Representations from Transformers (BERT)-Bidirectional Gated Recurrent Unit (BiGRU) to automatically identify NS in the reviews. The classifier based on the NS and BERT BiGRU multi-dimensional scales demonstrated a reasonably accurate recognition of the NS personality trait, with accuracy and F1 scores of 93.4% and 93.3%, respectively. This research also shows that by applying the deep learning model, the classifier based on multi-dimensional NS scales may provide good results. In comparison to other deep learning models of the same sort, the results also show that the BERT-BiGRU model performs at the highest level. It also demonstrates how personality characteristics may be automatically extracted from trip reviews using computational methods. Practically speaking, this research offers a thorough categorization framework for NS that can be used to recommendation and marketing systems in the travel and tourism sector.

DOI: 10.33545/26633582.2024.v6.i2b.132

Pages: 110-114 | Views: 520 | Downloads: 193

Download Full Article: Click Here