Feature Engineering & Exploratory Data Analysis

This repository contains techniques and notebooks for Feature Engineering (FE) and Exploratory Data Analysis (EDA). The repository is structured into two main directories:

• Feature Engineering: Covers various feature engineering techniques for handling missing values, encoding categorical data, balancing datasets, and more.
• EDA: Includes exploratory data analysis performed on different datasets.

Repository Structure

├── Feature Engineering/
│   ├── Data_encoding_Nominal_OHE.ipynb  # One-Hot Encoding for Nominal Data
│   ├── FE-handling_imbalanced_dataset.ipynb  # Handling Imbalanced Datasets
│   ├── FE-missing_values.ipynb  # Handling Missing Values
│   ├── Label_Ordinal_Encoding.ipynb  # Label & Ordinal Encoding
│   ├── Number_summary_&_Box_Plot.ipynb  # Summary Statistics & Box Plots
│   ├── Smote.ipynb  # Synthetic Minority Over-sampling Technique (SMOTE)
│   ├── Target_Guided_Ordinal.ipynb  # Target Guided Ordinal Encoding
│
├── EDA/
│   ├── flight_prices_prediction/  # EDA on Flight Prices Dataset
│   ├── wine-quality/  # EDA on Wine Quality Dataset
│   ├── google_playstore_dataset/  # EDA on Google Play Store Dataset

Feature Engineering Notebooks

1. Data Encoding Techniques

   • One-Hot Encoding (OHE)
   • Label Encoding
   • Ordinal Encoding (Target Guided)

2. Handling Missing Values

   • Imputation techniques
   • Handling missing data in different scenarios

3. Dealing with Imbalanced Datasets

   • SMOTE (Synthetic Minority Over-sampling Technique)
   • Other techniques for handling class imbalance

4. Exploring Numeric Features

   • Summary statistics
   • Box plots for outlier detection

Exploratory Data Analysis (EDA)

Each dataset in the EDA/ folder contains a detailed analysis covering:

• Data Cleaning
• Visualization
• Summary Statistics
• Insights & Trends

Usage

Clone the repository and run the Jupyter notebooks in your preferred environment.

git clone https://github.com/AlokTheDataGuy/Feature-Engineering-EDA.git

Requirements

Ensure you have the following dependencies installed:

pip install pandas numpy matplotlib seaborn scikit-learn imbalanced-learn

License

This project is open-source and available under the MIT License.