AN OPEN SOURCE HYBRID LEARNING OBJECT RECOMMENDER SYSTEM BASED ON EUROPEANA

Learning Object Repositories (LORs) are web-based digital libraries used for storing, distributing, discovering and retrieving learning objects. These repositories play a crucial role in the distribution of educational content by offering access to wide collections of learning objects. Besides offering learning object search services, many LORs offer additional features that allow users to create personal accounts, browse learning objects, bookmark learning objects, view and download metadata, or use educational tools. At present, one of the most well known LORs is Europeana, which provides access to millions of digitised materials from European museums, libraries, archives and multimedia collections. The main goal of Europeana is to provide access to Europe’s scientific and cultural heritage through a single access point.

The amount of resources available in LORs has grown significantly in recent years, causing users to have difficulty in finding relevant and quality content. One measure that LORs can adopt in order to face this problem is the use of recommender systems.

This paper presents EuropeanaRS: an open source hybrid learning object recommender system capable of generating recommendations of learning objects retrieved from Europeana. The paper describes the main features of the system, the supported use cases, the used knowledge sources, and the different steps of the recommendation process. An evaluation study with 20 participants was conducted in order to evaluate the accuracy, utility, user satisfaction and usability of EuropeanaRS. The results of this evaluation are also presented in this paper showing that EuropeanaRS had a satisfactory accuracy, was perceived as useful, achieved high user satisfaction, was easy to use, and was able to generate recommendations that significantly outperformed random recommendations.

Finally, the paper describes how EuropeanaRS was used in a tool developed in the Europeana Cloud project to facilitate the exploration of digitised newspapers retrieved from Europeana in order to provide this tool with recommendations. The results of an evaluation of this tool are also reported. These results show that EuropeanaRS can be successfully used to enrich tools by providing recommendations of learning objects retrieved from Europeana.