A component-based open-source framework for general-purpose recommender systems

Recommender systems constitute a new field that provides nowadays an important support for information search from huge amounts of data, since some of those information may be of interest of users but hard to be searched manually. According to forecasts, data production tends to grow more and more, which places recommender systems on the way to play a key role in mining those data. Technologies about to be adopted in large scale require tools to enable their mass production, and this fact is the key motivation of this work. In this paper we present our research in the field of component-based software engineering to design a framework for recommender systems. We have elicited the requirements that all recommender systems might address, and developed a vertical framework to support the development of such a sort of systems by designing components in accordance with those requirements. With the Idealize Recommendation Framework, we intend to make the development of recommender systems an easier, faster, and standardized process, as well as to perform the foundation of the technical terms to be used in the area.