Abstract
Good healthcare systems are crucial to the socio-economic growth of a country. In the current scenario, with a rapidly growing population, it is not always possible to find good hospitals nearby which has beds and doctors available, especially in times of emergencies. However, no reliable hospital recommendation system exists which compute these parameters and suggests hospitals. The proposed model utilizes neural networks to recommend hospitals based on hospital ratings, doctors available and distance to ensure that in time of emergencies, because crucial time is not wasted and the people can go to the best suited hospital. A comparative study among different neural network models were carried out in terms of accuracy and computing time. The most efficient model was chosen to be implemented in the final recommendation system. The dataset used to train the neural network models was the Hospital rating dataset provided by the Centers for Medicare and Medicaid services available on Kaggle. A real-time database has been made in Google Firebase, which contains information about hospitals like location, ratings, doctors and beds available. An attendance system for doctors has also been designed using Radio Frequency Identification (RFID) cards and NodeMCU. A web-application has also been designed where the user can obtain the name, address and phone number of the recommended hospital or hospitals.