Development of a Multi-Criteria Deep Learning Framework for Streamflow Modelling under Data-Scarce Conditions in Eswatini

1. Introduction 1.1 Background and Context 1.2 Problem Statement 1.3 Objectives of the Study 1.4 Structure of the Thesis 2. Literature Review 2.1 Streamflow Modelling Techniques 2.2 Deep Learning in Hydrology 2.3 Multi-Criteria Decision Making 2.4 Challenges in Data-Scarce Regions 2.5 Eswatini Hydrological Context 3. Methodology 3.1 Framework Design 3.2 Data Collection and Preprocessing 3.3 Model Development 3.4 Evaluation Criteria and Metrics 3.5 Software and Tools Used 4. Framework Development 4.1 Criteria Selection Justification 4.2 Model Training Process 4.3 Handling Data Scarcity 4.4 Integration of Multi-Criteria Analysis 4.5 Model Optimization Techniques 5. Case Study: Eswatini 5.1 Description of Study Area 5.2 Data Sources and Characteristics 5.3 Application of the Framework 5.4 Results and Discussion 6. Results 6.1 Model Performance Metrics 6.2 Comparative Analysis 6.3 Sensitivity Analysis 7. Discussion 7.1 Implications of Findings 7.2 Comparison with Existing Models 7.3 Limitations of the Study 8. Conclusion 8.1 Summary of Key Findings 8.2 Recommendations for Future Research 8.3 Conclusion Remarks