Social network based multimodal recommendation system for restaurants

Abstract

" In the modern world people do not tend to take meals only from their houses, they like to experience new places for eat. Because of that, we can see large number of restaurants around us and serves for their customers daily. Even people do not have the ability to go each after restaurants and find what is suite for their needs, but we can early predict the restaurants according to previous customer findings on the specific restaurant. Where we could definitely save time and money. For selecting such restaurants, people have identify the good and bad aspects of the restaurant before. However, with the upscale of technology, people got the ability to view the restaurants’ aspects before they choose. Reading people reviews and analysing is much time wasting process. To overcome that, author suggest a novel way to this restaurant domain. In this Project, the author has come up with a social media-based solution with multi-modal architecture, which could increase the potential of determining restaurants around the user. Both text and visual reviews gone through pre-processing and with their own classification techniques. Before the recommendation, the model has been trained using deep learning techniques. "