🍷 Wine Quality Data Cleaning and Imputation

This project focuses on data loading, exploratory data analysis (EDA), data cleaning, and missing value imputation using the Red Wine Quality dataset from the UCI Machine Learning Repository.

The main goal is to demonstrate good data preprocessing practices before applying Machine Learning models.

---

📌 Dataset

• Name: Wine Quality – Red Wine
• Source: UCI Machine Learning Repository
• Observations: 1,599 samples
• Features: 11 physicochemical variables + 1 target variable (quality)

---

🧰 Tools & Libraries

• Python 3
• Pandas
• NumPy
• Matplotlib
• Seaborn
• Scikit-learn
• SciPy

---

🔍 Project Overview

1. Data Loading

• Dataset loaded directly from the UCI repository
• Basic inspection of shape, data types, and sample rows

2. Exploratory Data Analysis (EDA)

• Descriptive statistics
• Distribution of wine quality
• Alcohol content distribution
• Correlation analysis

3. Data Cleaning (Key Steps)

• Handling missing values
• Standardizing column names
• Date format unification
• Outlier detection (IQR and z-score)
• Removing duplicates
• Numeric and text normalization
• Feature categorization

4. Missing Data Imputation

Artificial missing values (15%) were generated and handled using:

• Mean imputation
• Median imputation
• Mode imputation
• KNN imputation

The methods were compared, and mean imputation was selected for the final dataset.

---

🎯 Purpose

• Practice real-world data cleaning techniques
• Understand different imputation strategies
• Prepare data for Machine Learning models
• Build a solid Data Science portfolio project

---

📜 License

This project uses publicly available data from the UCI repository and is intended for educational purposes.