A Review of Healthcare Recommendation Systems Using Several Categories of Filtering and Machine Learning-Based Methods

Health-care Recommendation Systems (RSs) is a famous application of AI (artificial intelligence) that involves the investigators worldwide. Several ML methods are utilized to develop health-care RSs. Selecting the best ML method to give customers a service/product is an interesting task in health-care RSs. Currently, it is observed in the buying pattern of individuals from in-shop to online the outcome in accessibility of OI (online information) which is exponentially improved day by day. In shopping scenarios, these RSs must be capable of advising reliable solutions to the customers. The medical health-care RSs have to manage a large quantity of data by cleaning the reliable data depending on the data analysis, generated on the user behaviour made by the customers during their online terms. RSs can suggest suitable items to customers based on their interests and existing favourites, leading to improved sales. This paper described the overview of health-care RS, phases of RSs such as IC (information collection), learning, and recommender or predictor. This paper presents an overview of three main methods used to construct health-care RSs that are content-based, collaborative filtering (CF)-based and hybrid (HB)-based filtering methods and discuss several ML-based methods used in health-care RSs with their performance comparison. The ML-based methods include; MFM (matrix factorization model), CNN, MLP, SVD, etc.