Clothing_Store_Returns_ML_CaseStudy

Clothing Store Returns Prediction – ML Data Preparation

Prepared and transformed a 200-record bilingual (Arabic/English) clothing retail dataset for a Returns Prediction ML model. Key work: unified 8+ inconsistent text variants per column across Arabic and English, parsed 38 price values from text strings ('1057 جنيه' → float), imputed 50 missing values using Mode (Size) and Median (Rating), applied Label Encoding to 5 categorical columns, and MinMax Scaled Price/Quantity/Rating. Output: 5-sheet Excel workbook with Raw, Cleaned, Encoding Reference, ML-Ready, and Summary sheets.

· Target: Returns Classification (Random Forest / Logistic Regression)