E-learning is one of the most popular and beneficial uses of the Internet today. However, it presents a problem known as the "information overload". Current systems of e-learning stored too many contents: on various topics, for different ages, etc..; therefore users often are not sure how to find what they are needing between all the contents available. And this problem is even greater when considering that most current e-learning systems have simple search engines as tools to find resources. In this paper a new algorithm that incorporates recommendations of learning objects from a repository on an e-learning system is presented. At the end, a case study is performed to evaluate the proposal. Techniques of artificial intelligence are used to filter and customize information and recommended items.XII Workshop de Tecnología Informática Aplicada en EducaciónRed de Universidades con Carreras de Informática (RedUNCI)