💾 Data Cleaning using SQL

📄 Project Description

This project focuses on cleaning a real-world employee layoffs dataset sourced from Kaggle. The raw data was imported into MySQL, where several SQL operations were used to clean and prepare the dataset for analysis.

The objective was to resolve issues in the raw data such as:

• Duplicate records
• Inconsistent text formatting
• Missing or null values
• Incorrect data types

---

🛠️ Tools Used

• MySQL
• MySQL Workbench
• Kaggle (for data source)

---

🧹 Data Cleaning Workflow

✅ 1. Duplicate Removal

• Identified duplicate rows using ROW_NUMBER() with a CTE.
• Retained only the first occurrence, and deleted the rest.

---

✅ 2. Data Standardization

• Trimmed extra white spaces from the company column.
• Replaced inconsistent values like:
  • "united states" → "United States"
  • "crypto currency" → "crypto"
• Converted the date column from TEXT to DATE type using STR_TO_DATE() and ALTER TABLE.

---

✅ 3. Handling Null & Missing Values

• Replaced empty strings in industry with NULL.
• Used self-joins to fill in missing industry values based on the company.
• Removed rows where both total_laid_off and percentage_laid_off were null.

---

✅ 4. Final Cleanup

• Dropped helper columns like row_num used during the cleaning process.

---

📌 Outcome

The dataset is now clean, consistent, and ready for:

• Exploratory Data Analysis (EDA)
• Dashboard development (e.g., in Power BI, Tableau)
• Reporting and visualization

---