Diabetes Prediction using Machine Learning

---

Project Overview

This project predicts whether a patient is diabetic using Machine Learning techniques based on medical attributes and health indicators.

The project was developed during my Data Science Internship at Exposys Data Labs Pvt. Ltd. from 19 September 2022 to 18 October 2022.

The objective was to prepare the dataset, train machine learning models, analyze feature relationships, and predict diabetes with high accuracy.

---

Internship Details

Organization: Exposys Data Labs Pvt. Ltd.

Domain: Data Science

Duration: 19 September 2022 – 18 October 2022

Problem Statement

Diabetes is a chronic disease affecting people across all age groups. Early prediction can help individuals take preventive measures and improve their lifestyle.

Project Tasks

• Prepare the dataset
• Perform data preprocessing
• Train machine learning models
• Evaluate model performance
• Predict diabetes with high accuracy

---

Objectives

• Data preprocessing
• Exploratory Data Analysis (EDA)
• Feature correlation analysis
• Machine learning model training
• Disease prediction
• Accuracy evaluation

---

Technologies & Tools Used

• Python
• Pandas
• NumPy
• Scikit-Learn
• Matplotlib
• Jupyter Notebook

---

Dataset Features

The dataset contains the following medical attributes:

• Pregnancies
• Glucose
• Blood Pressure
• Skin Thickness
• Insulin
• BMI
• Diabetes Pedigree Function
• Age
• Outcome

---

Skills Demonstrated

• Data Cleaning
• Data Preprocessing
• Exploratory Data Analysis (EDA)
• Data Visualization
• Machine Learning
• Logistic Regression
• Classification Models
• Python Programming
• Statistical Analysis

---

Repository Structure

Diabetes-Prediction-Model/
│
├── dataset/
│   └── diabetes.csv
│
├── source-code/
│   ├── diabetes_prediction.py
│   ├── main.c
│   └── read csv.cbp
│
├── screenshots/
│   ├── first-5-records.PNG
│   ├── feature-correlations-output.PNG
│   ├── correlation-heatmap.PNG
│   ├── diabetic-vs-non-diabetic-barplot.PNG
│   ├── average-age-analysis.PNG
│   ├── logistic-regression-weights.PNG
│   └── terminal-output.JPG
│
├── report/
│   └── project-report.docx
│
├── presentation/
│   └── diabetes-presentation.pptx
│
├── certificates/
│   ├── exposys-data-science-internship-certificate.pdf
│   ├── simplilearn-introduction-to-data-science-certificate.pdf
│   └── basics-of-machine-learning-certificate.pdf
│
├── README.md
└── LICENSE

---

Project Screenshots

First 5 Records in Diabetes Dataset

Feature Correlation Output

Correlation Heatmap

Diabetic vs Non-Diabetic Distribution

Average Age Analysis

Logistic Regression Feature Weights

Terminal Output

Project Workflow

1. Load Dataset
2. Data Cleaning & Preprocessing
3. Exploratory Data Analysis (EDA)
4. Correlation Analysis
5. Feature Selection
6. Model Training
7. Logistic Regression Implementation
8. Accuracy Evaluation
9. Diabetes Prediction

---

Project Results

The model analyzes patient health data and predicts the likelihood of diabetes based on medical indicators.

Key Activities

• Dataset preprocessing
• Feature correlation analysis
• Data visualization
• Model training
• Logistic Regression implementation
• Prediction generation
• Accuracy evaluation

---

Source Code

Python Implementation

• diabetes_prediction.py

Additional Files

• main.c

• read csv.cbp

---

Project Documentation

Report

• Project Report

Presentation

• Project Presentation

---

Certifications

Internship Certificate

• Exposys Data Science Internship Certificate

Additional Learning Certifications

• Introduction to Data Science – Simplilearn

• Basics of Machine Learning Algorithms – UniAthena & Cambridge International Qualifications

---

Learning Outcomes

• Data preprocessing techniques
• Exploratory Data Analysis (EDA)
• Correlation analysis
• Machine learning fundamentals
• Logistic Regression implementation
• Model evaluation techniques
• Healthcare data analysis

---

Future Improvements

• Deploy as a Web Application
• Add Multiple Machine Learning Algorithms
• Improve Prediction Accuracy
• Create Interactive Dashboard
• Real-Time Prediction Interface
• Streamlit Deployment
• Model Comparison Dashboard

---

License

This project is licensed under the MIT License.

---

Author

Arpeet Bhaisare

B.Tech – Electronics & Communication Engineering Indian Institute of Information Technology (IIIT) Bhopal

Connect With Me

LinkedIn: https://www.linkedin.com/in/arpeet-bhaisare-a02a061ba

GitHub: https://github.com/Arpeet-Bhaisare

---

⭐ If you found this project useful, consider giving it a star.