📊 Exploratory Data Analysis (EDA) Projects

This repository contains Exploratory Data Analysis (EDA) workflows for three datasets: Ford Used Cars, Heart Disease Diagnostic Data, and Medical Insurance Charges.

🛠️ Environment & Prerequisites

• Python Version: Python 3.14+
• Environment: Jupyter Notebook / VS Code

Installed Libraries (requirements.txt)

• pandas - Data manipulation and analysis
• numpy - Numerical computing
• matplotlib - Static visualizations
• seaborn - Statistical data visualization
• scikit-learn - Machine learning preprocessing & baseline modeling
• jupyter - Interactive notebook interface

---

🔬 Exploratory Data Analysis Workflow Steps

Across all three notebooks, the EDA follows a standardized 6-step workflow:

1. Data Ingestion & Initial Inspection (.head(), .info(), .describe())

   • Loaded raw CSV files into Pandas DataFrames.
   • Examined data types, dataset shapes, and summary statistics.

2. Data Cleaning & Preprocessing

   • Handled missing/null values and identified duplicate records.
   • Fixed column typos and ensured appropriate datatypes for features.

3. Univariate Analysis

   • Plotted feature distributions using histograms, box plots, and count plots.
   • Identified and managed extreme outliers in numerical columns.

4. Bivariate & Multivariate Analysis

   • Created heatmaps to observe correlation matrices between target and independent variables.
   • Generated pair plots and scatter plots to inspect feature interactions.

5. Categorical Feature Encoding

   • Processed categorical variables using One-Hot and Label Encoding techniques for model readiness.

6. Insights & Feature Summary

   • Extracted key business and clinical takeaways for reporting and future predictive modeling.

---

📂 Notebook Summaries

🚗 ford_car_datasate.ipynb

• Dataset: ford.csv
• Focus: Analyzed mileage, registration year, engine size, and fuel type correlations against car resale values.

❤️ heart EDA .ipynb

• Dataset: heart.csv
• Focus: Evaluated heart disease risk factors such as maximum heart rate, chest pain type, blood pressure, and age distributions.

🛡️ insurance EDA & Machin learning practice .ipynb

• Dataset: insurance.csv
• Focus: Examined primary drivers of medical costs (notably smoking status and BMI interactions) and built initial ML regression pipelines.