A MACHINE LEARNING-BASED FEMALE BODY SHAPE CLASSIFIER

This research presents the development of a machine learning-based female body shape prediction and style recommender model aimed at improving personalized fashion recommendations. The study focuses on classifying female body shapes into major categories such as Hourglass, Rectangle, Apple, Pear, and Inverted Triangle using key body measurements including shoulder width, bust circumference, waist circumference, and hip circumference. A machine learning model was developed using Python-based tools and trained on a dataset collected from professional fashion designers’ measurement records. Data preprocessing techniques such as cleaning, normalization, and one-hot encoding were applied before model training. The developed system predicts body shape and provides personalized styling recommendations, particularly for blouse and neckline designs. Findings from the study reveal that the model achieved satisfactory prediction accuracy and demonstrates the viability of integrating machine learning into personalized fashion recommendation systems. The study concludes that machine learning can significantly enhance fashion personalization, user confidence, and customer satisfaction within the fashion industry.