Predictive Analysis on PIMA Indian Diabetes Dataset

Project Overview

This project involves performing predictive analysis on the PIMA Indian Diabetes dataset to identify whether a patient is likely to develop diabetes based on several medical attributes. Using various machine learning algorithms, the analysis aims to improve early detection and assist healthcare professionals in decision-making.

Project Details

• Name: Ashika Jain
• Company: CODTECH IT SOLUTIONS
• ID: CT08FDI
• Domain: Data Analytics
• Duration: December 2024 to January 2025

Dataset Description

The PIMA Indian Diabetes dataset contains medical data for women of Pima Indian heritage aged 21 and older. The dataset includes the following features:

1. Pregnancies: Number of times pregnant.
2. Glucose: Plasma glucose concentration after 2 hours in an oral glucose tolerance test.
3. BloodPressure: Diastolic blood pressure (mm Hg).
4. SkinThickness: Triceps skin fold thickness (mm).
5. Insulin: 2-hour serum insulin (mu U/ml).
6. BMI: Body mass index (weight in kg/(height in m)^2).
7. DiabetesPedigreeFunction: Diabetes pedigree function (a measure of hereditary diabetes likelihood).
8. Age: Age (years).
9. Outcome: Binary variable indicating the presence (1) or absence (0) of diabetes.

Objectives

• To preprocess and clean the dataset for analysis.
• To perform exploratory data analysis (EDA) to identify patterns and correlations.
• To build and evaluate machine learning models for diabetes prediction.
• To provide insights and recommendations based on the model's outcomes.

Tools and Technologies

• Programming Language: Python
• Libraries: Pandas, NumPy, Matplotlib, Seaborn, Scikit-learn
• Environment: Jupyter Notebook/Google Colab

Workflow

1. Data Preprocessing:

   • Handling missing values.
   • Normalizing and scaling data.
   • Encoding categorical variables (if applicable).

2. Exploratory Data Analysis:

   • Visualization of feature distributions.
   • Correlation analysis.
   • Identifying outliers and anomalies.

3. Model Development:

   • Splitting data into training and testing sets.
   • Training machine learning models (e.g., Logistic Regression, Decision Trees, Random Forest, Support Vector Machine).
   • Hyperparameter tuning.

4. Model Evaluation:

   • Metrics: Accuracy, Precision, Recall, F1-score, ROC-AUC.
   • Comparing model performances.

5. Insights and Recommendations:

   • Highlighting key features contributing to diabetes prediction.
   • Proposing actionable strategies for healthcare intervention.

Results

Conclusion

The predictive analysis demonstrated that the PIMA Indian Diabetes dataset could effectively be used to predict diabetes using machine learning models. By leveraging features such as glucose levels, BMI, and age, the model provides reliable predictions, which can aid in proactive healthcare strategies.