Recommendation System For Personlised Customer Engagement Using Resnet

As the fashion industry and e-commerce websites develops rapidly, personalized customer engagement becomes the key for improving user experience and sales. Traditional recommendation systems, which were based on users' history and collaborative filtering techniques, were not able to capture the current fashion trends and the recent interests of customers. To bridge the gap, this paper presents a real-time AI-powered fashion recommendation system that utilizes technologies such as computer vision and natural language processing (NLP) to recommend tailored outfits. It uses a Res Net model to extract the fashion features, which are important features of clothing, such as category, pattern, sleeve type, length, neckline, material, etc., .A BERT-based NLP model is designed to provide text-based suggestions from these features, by doing a real-time trend analysis and outfit pairing. The model is trained on the deep fashion dataset, which provides context-aware suggestions, enhancing online and offline shopping experiences. With the advancement of computer vision and NLP, this model improves the relevance and accuracy of suggestions, enhancing the customer satisfaction and engagement.