M.A. Thesis:

Development of an algorithm using genetic algorithms for recommender systems

Today, due to the growing Internet and massive amounts of information, we need systems to recommend the most appropriate services and products to the user, the systems that perform this task are called advisory systems. The advocate systems, using certain algorithms and methods, try to identify the most relevant items, such as data, information, and goods, and propose the closest item to the user's taste. Advisory systems are a branch of information retrieval and information systems that, by identifying the interests and needs of the user, help them access the information or services they want through a massive selection of choices. The lack of sufficient data and the dispersion of data are among the challenges of the advisory systems. In this study, we first evaluated the data with the preprocessing process so that the less valuable data were eliminated or minimized, then we proposed the most suitable items using a binary clustering method and genetic algorithm. The results of the simulation show that the measurement factor, which is one of the fundamental components of the evaluation of these types of systems, has been improved.

Keywords: Recommender systems, social networks, genetic algorithms