| Titre : | Smart Predictive Agriculture Based on Data Science |
| Auteurs : | M’HAMED MANCER, Auteur ; Sadek Labib Terrissa, Directeur de thèse |
| Type de document : | Monographie imprimée |
| Editeur : | Biskra [Algérie] : Faculté des Sciences Exactes et des Sciences de la Nature et de la Vie, Université Mohamed Khider, 2025 |
| Format : | 1 vol. (94 p.) / ill., couv. ill. en coul / 30 cm |
| Langues: | Anglais |
| Langues originales: | Anglais |
| Mots-clés: | Smart Predictive Agriculture ; Crop Selection ; Interpretable Machine Learning ; SHAP ; Tomato Yield Prediction ; Ensemble learning, Blockchain ; Data Integrity ; Decision Support Systems. |
| Résumé : |
Smart agriculture integrates digital technologies, sensors, the Internet of Things, big data, and artificial intelligence to transform traditional farming into precision-oriented and data driven systems. These systems aim to improve productivity while making better use of resources. At the beginning of each growing season, farmers must make decisions that guide the success of the entire production cycle. The most important of these choices is deciding which crops to plant and how to divide land among them. This choice influences all later activities, such as planning the planting schedule, preparing the soil, and rganizing the use of inputs. Because of its importance, crop selection is often described as the first step in farm planning. The first contribution of this thesis responds to this problem by introducing an interpretable crop selection system. The system integrates SHAP-based explanations to show how soil properties and climate conditions affect each recommendation. It combines strong predictive ability with clear explanations, offering a practical tool that farmers and advisors can use with greater trust. After the crop has been chosen, the next important question is “how much to expect.” Accurate yield forecasting allows farmers to organize inputs, schedule labor, manage uncertainty, and prepare for market activities. The second contribution of this thesis addresses this by designing a stacked ensemble learning framework, developed with greenhouse tomato production as a case study. It delivers accurate daily yield forecasts and achieves better results than standard regression methods, providing a reliable decision-support tool for greenhouse management. Since both crop selection and yield forecasting depend on the quality of agricultural data, the third contribution focuses on how this data can be kept secure, reliable, and trustworthy. To achieve this, a blockchain-based approach is proposed that integrates encryption, distributed file storage, and smart contracts. The approach ensures data traceability, confidentiality, and tamper-resistance. |
| Sommaire : |
Abstract i Acknowledgements iv List of Abbreviations xv 1 General introduction 1 1.1 Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.2 Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.3 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.4 Thesis Structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 2 Preliminaries and Basic Concepts 8 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2.2 Data Science Fundamentals . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2.2.1 Definition and Scope . . . . . . . . . . . . . . . . . . . . . . . . . . 8 2.2.2 Data Science Lifecycle . . . . . . . . . . . . . . . . . . . . . . . . . 10 2.2.3 Data Preprocessing & Feature Engineering . . . . . . . . . . . . . . 11 2.2.4 Exploratory Data Analysis & Visualization . . . . . . . . . . . . . 14 2.3 Machine Learning and Deep Learning . . . . . . . . . . . . . . . . . . . . . 17 2.3.1 Machine Learning Basics . . . . . . . . . . . . . . . . . . . . . . . . 17 2.3.2 Methodologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 2.3.3 Model Evaluation Metrics . . . . . . . . . . . . . . . . . . . . . . . 20 2.3.4 Deep Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 2.4 Interpretable and Explainable AI (XAI) . . . . . . . . . . . . . . . . . . . 25 2.4.1 Importance Across Domains . . . . . . . . . . . . . . . . . . . . . . 25 2.4.2 Foundations of Explainable AI . . . . . . . . . . . . . . . . . . . . 26 2.4.3 Post-hoc Explanations . . . . . . . . . . . . . . . . . . . . . . . . . 26 2.4.4 SHAP for Model Interpretability . . . . . . . . . . . . . . . . . . . 28 2.4.5 Visualization and User Interfaces . . . . . . . . . . . . . . . . . . . 28 2.5 Blockchain Technology for Data Security . . . . . . . . . . . . . . . . . . . 30 2.5.1 Fundamentals of Blockchain . . . . . . . . . . . . . . . . . . . . . . 30 2.5.2 Types of Blockchain Networks . . . . . . . . . . . . . . . . . . . . . 30 2.5.3 Security and Privacy Features . . . . . . . . . . . . . . . . . . . . . 31 2.6 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 3 Smart Agriculture: State of the Art 33 3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 3.2 Traditional Agriculture: Challenges and Limitations . . . . . . . . . . . . . 34 3.2.1 The Global Importance of Agriculture . . . . . . . . . . . . . . . . 34 3.2.2 Key Types of Agricultural Challenges . . . . . . . . . . . . . . . . 37 3.2.3 Traditional Agricultural Practices and Their Limitations . . . . . . 39 3.3 AI-Driven Transformation of Agriculture . . . . . . . . . . . . . . . . . . . 40 3.3.1 From Traditional Practices to Intelligent Systems . . . . . . . . . . 40 3.3.2 Core AI Technologies in Smart Agriculture . . . . . . . . . . . . . . 41 3.3.3 Impact on Productivity and Sustainability . . . . . . . . . . . . . . 43 3.4 Literature Review on Crop Selection Systems . . . . . . . . . . . . . . . . 44 3.4.1 Descriptive Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . 44 3.4.2 Related Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 3.4.3 Research Gaps and Contribution . . . . . . . . . . . . . . . . . . . 52 3.5 Literature Review on Data-Driven Crop Yield Prediction . . . . . . . . . . 53 3.6 Literature Review on Blockchain Applications for Data Security . . . . . . 55 3.7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 4 Contribution 1: Interpretable Crop Selection for Optimized Farming Decisions 59 4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 4.2 Methodology Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 4.3 Exploratory Data Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 4.3.1 Data Acquisition and Characteristics . . . . . . . . . . . . . . . . . 62 4.3.2 Univariate Analysis and Distribution Visualization . . . . . . . . . 63 4.3.3 Bivariate Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 4.3.4 EDA-Driven Strategy . . . . . . . . . . . . . . . . . . . . . . . . . 70 4.4 Data Preprocessing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 4.4.1 Data Cleaning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73 4.4.2 Feature Scaling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 4.4.3 Categorical Encoding . . . . . . . . . . . . . . . . . . . . . . . . . . 76 4.4.4 Feature Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 4.4.5 Data Augmentation . . . . . . . . . . . . . . . . . . . . . . . . . . 77 4.5 Assessment of Data Preprocessing . . . . . . . . . . . . . . . . . . . . . . . 78 4.5.1 Experimental Design . . . . . . . . . . . . . . . . . . . . . . . . . . 78 4.5.2 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 4.6 Development and Workflow of the CS-AdaRF-SHAP Model . . . . . . . . 83 4.6.1 CS-AdaRF Development and Optimization . . . . . . . . . . . . . . 84 4.6.2 SHAP-Based Interpretability . . . . . . . . . . . . . . . . . . . . . 88 4.7 Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 4.7.1 Evaluation of the CS-AdaRF Model . . . . . . . . . . . . . . . . . 90 4.7.2 Assessment of Model Interpretability . . . . . . . . . . . . . . . . . 94 4.8 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 4.9 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 5 Contribution 2: Data-Driven Crop Yield Prediction 100 5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 5.2 Materials and Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 5.2.1 System Architecture Overview . . . . . . . . . . . . . . . . . . . . . 101 5.2.2 Dataset Description . . . . . . . . . . . . . . . . . . . . . . . . . . 102 5.2.3 Data Preprocessing . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 5.3 Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106 5.3.1 Predictive Performance Comparison . . . . . . . . . . . . . . . . . 107 5.3.2 Error Analysis and Robustness . . . . . . . . . . . . . . . . . . . . 108 5.3.3 Alignment of Predicted and Actual Yields . . . . . . . . . . . . . . 109 5.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 6 Contribution 3: Blockchain-Based Approach to Securing Data in Smart Agriculture 111 6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111 6.2 Proposed Solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 6.2.1 Architecture of the Proposed Approach . . . . . . . . . . . . . . . . 113 6.2.2 Structure of Blocks and Transactions . . . . . . . . . . . . . . . . . 114 6.3 Implementation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 6.3.1 Development Environment and Tools . . . . . . . . . . . . . . . . . 116 6.3.2 Smart Contract Deployment and Data Transactions . . . . . . . . . 116 6.3.3 Data Encryption and Secure Storage . . . . . . . . . . . . . . . . . 117 6.3.4 Performance and Security Analysis . . . . . . . . . . . . . . . . . . 118 6.4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 7 General Conclusion and Perspectives 120 7.1 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120 7.2 Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122 List of Publications 124 References 125 |
| Type de document : | Mémoire master |
Disponibilité (1)
| Cote | Support | Localisation | Statut |
|---|---|---|---|
| TINF/209 | Théses de doctorat | bibliothèque sciences exactes | Consultable |
