MFMwapashua H. Fujo

Mwapashua H. Fujo

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Mwapashua H. Fujo has a total of 1 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Predicting Tourist Decision-Making Under Data Scarcity: A Hybrid Machine Learning Framework for Digital Marketplaces

    by Mwapashua H. Fujo, Angel Gabriel Meela

    Understanding and predicting consumer decision-making in digital environments remains a fundamental challenge, particularly in contexts characterized by limited and heterogeneous data. While advances in Machine Learning have enabled significant improvements in behavioral prediction, most existing approaches rely on large-scale datasets, limiting their applicability in real-world settings such as small and medium-sized enterprises (SMEs) and emerging markets. This study proposes a hybrid machine learning framework for predicting tourist decision making under data scarcity conditions in digital marketplaces. The framework integrates ensemble learning and neural modeling to capture both static and temporal behavioral patterns, while incorporating contextual and interaction-based features. Experimental evaluation was conducted using a real-world dataset of 10,000 user sessions collected from a tourism platform in Tanzania. To assess robustness, controlled experiments were performed under varying data availability scenarios. Results demonstrate that the proposed hybrid model consistently outperforms baseline models, achieving superior performance across multiple evaluation metrics, including accuracy, F1-score, and AUC-ROC. Notably, the model maintains strong predictive capability even when trained on reduced datasets, highlighting its robustness in data-constrained environments. Further more, interpretability analysis using Explainable AI reveals key behavioral and contextual factors influencing user decisions. The findings contribute to bridging the gap between advanced predictive analytics and real-world applicability by providing a scalable and adaptable framework for AI-driven decision modeling in low-data environments. The proposed approach has broad implications for digital marketplaces, particularly in emerging economies, where data limitations remain a critical barrier to the adoption of intelligent systems.