Liver Disease Prediction : using Ensemble Learning

The Liver Disease Prediction Model is a powerful ensemble learning-based system that accurately predicts the presence of liver disease in patients. The models success is attributed to its use of various classifier models working together, leading to enhanced accuracy and robustness. Users can interact with the Liver Disease Prediction web application to input their medical attributes, including age, gender, and various blood test results. Upon submission, the application swiftly processes the data through the ensemble model, providing predictions and performance metrics.

One of the standout features of the system is its exceptional accuracy of 100%. While this might raise concerns about potential overfitting or data leakage, the developers acknowledge these issues and encourage users to validate the models performance on unseen data before deploying it in real-world scenarios. To foster a better understanding of the models capabilities, the web application allows users to compare its results with other well-known classifier models.

To ensure the models reliability, the Liver Disease Prediction Model is trained on a comprehensive dataset named Liver.csv. Prior to training, data preprocessing steps, such as handling missing values and encoding categorical variables, are applied. This ensures that the model learns from high-quality data and produces accurate predictions.

• Ensemble Learning: The model uses a combination of various classifier models to enhance accuracy and robustness.
• Web Application: The project offers a user-friendly web application built with Streamlit, enabling users to interact with the prediction model easily.
• 100% Accuracy: The model boasts an impressive 100% accuracy in predicting liver disease, but users are reminded to perform thorough evaluations before deployment.
• Model Comparison: Users can compare the ensemble models performance with other selected classifiers, including Random Forest, Naïve Bayes, Logistic Regression, and more.
• Dataset: The model is trained on the Liver.csv dataset, containing essential input features and corresponding target labels.
• Data Preprocessing: The dataset undergoes preprocessing steps, such as handling missing values and encoding categorical variables, to ensure reliable model training.
• Easy Installation: Instructions are provided for cloning the repository and installing necessary libraries to run the application locally or contribute to the project.
• Open for Contributions: The project is open for contributions, and users can submit pull requests or open issues for feedback and improvements.