Court Judgments Prediction using Machine Learning

Supreme Court of Nigeria (SCN) appeal-case outcome prediction with interpretable ML workflows.

Court Judgments Prediction — README

This project predicts judicial outcomes for Supreme Court of Nigeria appeal cases using supervised machine learning models and compares their performance with visual diagnostics and confusion matrices.

Table of Contents

• 🚀 Project intro
• 📁 Project structure
• 📊 Dataset
• 🧠 Modeling approach
• 🤖 Models evaluated
• 📈 Outputs and artifacts
• ⚙️ How to run
• 📄 License

🚀 Project intro

The repository demonstrates an end-to-end ML workflow for legal outcome prediction:

• Data profiling and exploratory analysis of SCN appeal data
• Feature engineering and preprocessing (including encoding/scaling)
• Training and comparison of multiple classification algorithms
• Performance reporting with confusion matrices and plots

📁 Project structure

Court-Judgments-Prediction-using-Machine-Learning/
├── CSE 445.ipynb
├── scn_appeal_cases_data.csv
├── datasetprofiling.html
├── Dataset description.pdf
├── IMAGES/
│   ├── Confusion Matrix for LogisticRegression().png
│   ├── Confusion Matrix for SVC().png
│   ├── Feature_corelation_rc_heatmap_plot.png
│   └── ...
├── LICENSE
└── README.md

📊 Dataset

• Source: Primsol Law Pavilion archive (distributed via Mendeley)
• Records: 5,585 appeal cases
• Scope: Criminal and civil appeal matters from the Supreme Court of Nigeria
• Dataset link: https://data.mendeley.com/datasets/ky6zfyf669/1

🧠 Modeling approach

The notebook performs:

• Data cleaning and exploratory visual analysis
• Feature transformation with LabelEncoder and scaling tools where required
• Train/test split and model training
• Evaluation using accuracy metrics, classification reports, and confusion matrices

🤖 Models evaluated

Primary and baseline models included in the notebook:

• DummyClassifier
• DecisionTreeClassifier
• KNeighborsClassifier
• LogisticRegression
• SVC / OneVsRestClassifier
• GaussianNB
• MLPClassifier
• SGDClassifier
• Additional ensemble experiments (AdaBoost, Bagging, RandomForest, etc.)

📈 Outputs and artifacts

• datasetprofiling.html contains automated dataset profiling
• IMAGES/ contains confusion matrices and EDA plots
• CSE 445.ipynb contains the full workflow and model comparisons

⚙️ How to run

1. Install Python 3 and Jupyter.
2. Install dependencies used in the notebook:

pip install pandas numpy matplotlib seaborn scikit-learn pandas-profiling jupyter

3. Open and run:

jupyter notebook "CSE 445.ipynb"

📄 License

This project is licensed under the MIT License. See the LICENSE file for details.