This repository contains a machine learning project for predicting diabetes based on demographic and clinical data. The project leverages various machine learning models and techniques to develop an accurate predictive model. This was our undergraduate project for the CSE422 course at BRAC University.
Diabetes is becoming a major public health concern around the world, necessitating the development of reliable prediction models to aid in early detection and intervention. Our project aims to create a prediction model for diabetes detection using Machine Learning methods.
The dataset used for this project contains demographic and clinical features, including age, gender, BMI, HbA1c level, blood glucose level, hypertension, heart disease, and smoking history. The target variable is binary, indicating the presence or absence of diabetes.
Dataset Source: Kaggle Diabetes Prediction Dataset
We have used 4 ML Models to train our dataset:
- k-Nearest Neighbors (KNN)
- Support Vector Machine (SVM)
- Decision Tree
- Random Forest
While assessing the SVM model, we experimented with kernels including Linear, RBF, Poly, and Sigmoid.
- Notebook: Jupyter notebook
Diabetes Prediction Project.ipynbcontaining the project code and analysis. - Dataset: CSV file
Diabetes Prediction Dataset.csvcontaining the dataset used for training and testing. - Report: Project report
Project Report.pdfsummarizing the project objectives, methodology, results, and conclusions.
To run the project code, follow these steps:
-
Clone the repository:
git clone https://github.com/niloyahsan1/Diabetes-Prediction-ML-Project.git
-
Navigate to the project directory:
cd Diabetes-Prediction-ML-Project -
Open and run the Jupyter Notebook (
Diabetes_Prediction_Project.ipynb) to execute the project code.
The project requires the following Python libraries:
- pandas
- numpy
- matplotlib
- seaborn
- scikit-learn
You can install these dependencies using pip:
pip install pandas numpy matplotlib seaborn scikit-learn- Chaity Rani Ghosh (@chaityrg)