Virtual Smart Canteen System

The proposed work targets to solve the issues faced by students in a university canteen by integrating a mobile application for online food ordering, together with an application for vacant seat detection for hassle-free seating space finding, food recommendation and sentiment analysis features making food ordering process in a university canteen more convenient and efficient. Our proposed system consist of three main modules which are Vacant seat detection module, Recommendation module and Feedback analysis module. We have used TensorFlow detection model for our vacant seat detection. We’ve used this model to detect people in canteen and calculate the available seats. This helps students, faculty and staff to be aware of the estimated number of canteenâs customers in real-time. To implement recommendations module, our system uses item-based collaborative filtering. This module give five similar food item based on the similarity of food items. For feedback analysis module, we analyzed the sentiment of the comments received on a particular food item. A positive review is defined as a score of 4 or 5, negative rating of 1 or 2 can be considered and review with a rating of 3 is considered neutral. This is used to know the overall rating of food. All three module combined can definitely help students/faculty to save their time as well as help canteen staff members to manage everything for effectively.