Many-objective optimization meets recommendation systems: A food recommendation scenario

Due to the ever-increasing amount of various information provided by internet, recommendation systems are now used in a large number fields as efficient tools get rid overload. The content-based, collaborative-based and hybrid methods three classical techniques, whereas not all real-world problems (e.g. food problem) can be best addressed such techniques. This paper is devoted solving problem based on many-objective optimization (MaOO). A novel approach proposed transforming original into an MaOO one that contains four different objectives, i.e., user preferences, nutritional values, dietary diversity, diet patterns. experimental results demonstrate designed provides more balanced way recommending than only consider individuals’ preferences.