📊 Outlier Detection and Data Splitting using IQR

A Python CLI tool to detect outliers using IQR, split datasets, print summary stats, and save results as CSV.

📌 Project Objective

This Python script:

• Accepts a CSV file and a target column as input
• Performs Interquartile Range (IQR) based outlier detection
• Splits the dataset into two parts: outliers and non-outliers
• Calculates and prints summary statistics (mean, median, std, IQR) to the console
• Saves both subsets as separate CSV files (outliers.csv, non_outliers.csv)

⚠️ Note: No plotting or HTML reports are included. The output is purely statistical and text-based.

---

🧠 Skills Demonstrated

• CLI-based data preprocessing and outlier detection
• Efficient reading, writing, and filtering of CSV data using Python
• Statistical analysis using the IQR method
• File I/O operations with pandas
• Modular code design with argument handling and clean output

---

🧰 Libraries Used

• pandas: for data manipulation and analysis
• numpy: for numerical computations
• sys: to handle command-line arguments

---

🚀 How It Works

📁 Steps Performed

1. Import Required Libraries
2. Accept Command-Line Arguments
   python outlier_split.py <input_file.csv> <column_index>
3. Read the Input CSV
4. Extract the Target Column
5. Perform IQR-Based Outlier Detection
6. Calculate and Print Summary Statistics
   (Mean, Median, Std Dev, IQR, Min, Max)
7. Save Subsets as CSV Files
   • outliers.csv
   • non_outliers.csv

---

⚙️ Usage & Example Output

python outlier_split.py data.csv 2

Example Output:

Original Data Summary:
Mean: 57.2 | Median: 55.0 | Std Dev: 12.6 | IQR: 15.5

Outlier Subset Summary:
Mean: 98.3 | Count: 6

Non-Outlier Subset Summary:
Mean: 53.1 | Count: 94