The project was to develop an online portal for shoppers that provides recom- mendations based on similarities in the design patterns, taking a step beyond the usual searching on price, material, color etc. The process started with gathering data from multiple online resources by scraping the websites of dierent online stores within the city. Image Processing was the next crucial part as texture features had to be extracted from the images of printed clothes to recognize their patterns. Machine Learning came in the form of unsupervised learning where clustering techniques were explored on a sample data set. The best technique was chosen based on highest quality of clusters formed. K-means, Self Organizing Maps, and Adaptive Resonance Theory were the clustering algorithms applied on the image features extracted using Haralick fea- tures extraction technique and also on features from Auto encoders. The results from Haralick features undergone through Self Organising Maps clustering were used to nally generate suggestions for similar items on the website