Skip to content

Diabetes prediction problem (Binary Classification) using 5 ML models - Beginner Level

License

Notifications You must be signed in to change notification settings

SivadithiyanOfcl/pima-diabetes

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

pima-diabetes

Please leave a upvote and drop a comment!

Diabetes prediction problem (Binary Classification) using 5 ML models - Beginner Level

Diabetes Prediction with Pima Indian Dataset

📊 This project aims to predict diabetes using the Pima Indian Diabetes Dataset through binary classification.

Notebook

Please visit the notebook posted on kaggle for the best experience

Contents of the notebook

  • 📊 Data splitting for training and testing
  • 🛠️ Data preprocessing to prepare the data for modeling
  • 🔍 Simple EDA (Exploratory Data Analysis) to understand the dataset
  • ⚖️ Class balancing using SMOTE to address imbalanced data
  • 🎛️ Parameter tuning using GridSearchCV to optimize model performance
  • 🤖 Training 5 different classification models to predict diabetes
  • 💾 Exporting the trained models for future use

If you can't see the notebook preview on github, open the pdf. It should load most of the times.

Overview

🤔 Diabetes is a prevalent health concern, and early prediction can aid in timely intervention and management. This project utilizes machine learning algorithms to predict the onset of diabetes based on various health parameters.

Dataset

📈 The dataset contains health-related information such as glucose levels, blood pressure, and BMI of individuals from the Pima Native American tribe. It's widely used in research for diabetes prediction.

Approach

🔍 Exploration of multiple machine learning algorithms such as Logistic Regression, Decision Trees, Support Vector Machines, K-Nearest Neighbors, and Random Forests to find the best model for diabetes prediction.

Usage

💻 To use this project:

  • Clone the repository.
  • Run the notebook or scripts to train and evaluate the models. (requires any code editor with appropriate libraries installed or you can load it to kaggle)
  • Analyze the results and fine-tune/download the models as needed.

Results

📈 Evaluation the models are based on accuracy and F1 score, providing insights into their performance and suitability for diabetes prediction.

About

Diabetes prediction problem (Binary Classification) using 5 ML models - Beginner Level

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published