Abstract :

The Rapid Expansion Of Digital Platforms In The Tourism Sector Has Resulted In A Massive Accumulation Of Unstructured User-generated Reviews, Making It Challenging To Extract Meaningful Insights For Accurate Demand Prediction. Conventional Classifiers Such As Linear Discriminant Analysis (LDA), Histogram Gradient Boosting (HGB), And Quadratic Discriminant Analysis (QDA) Are Limited In Capturing Deep Contextual Relationships And Struggle To Perform Effectively Under Data Imbalance Conditions. To Address These Limitations, This Study Proposes A Hybrid Framework Named GPS Tourism, Which Integrates Advanced Language Modeling With Robust Data Balancing Techniques. The Framework Leverages Google Pathways Language Model (PaLM) To Convert Raw Textual Reviews Into 768- Dimensional Semantic Embeddings, Enabling A Comprehensive Understanding Of Contextual Sentiment. To Mitigate Class Imbalance, The Synthetic Minority Over-sampling Technique (SMOTE) Is Applied, Ensuring Balanced Representation Of Minority Classes. The Final Classification Is Performed Using A Sparse Linear Integer Model (SLIM), Structured As An Ensemble Of Oblique Decision Trees To Enhance Both Interpretability And Prediction Accuracy. Experimental Results Demonstrate That The Proposed Model Outperforms Baseline Approaches. The Framework Enables Precise Demand Forecasting, Supporting Datadriven Decision-making For Resource Allocation And Targeted Marketing In The Tourism Industry.

Published:

09-4-2026

Issue:

Vol. 26 No. 4 (2026)

Page Nos:

2046-2056

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite