Abstract-This paper discusses the role of context in making recommender systems for digital libraries. It first introduces different types of recommender systems then discusses the role of context in making recommendations for users searching in digital libraries. This study highlights the important role of context and the need for more investigation on context-aware recommender system since exploiting contextual information in recommender systems is an effective approach to create more accurate and relevant recommendations. The article stresses the need for qualitative research on the subject in order to find out the elements that constitute the context from the users' viewpoint and the way a context-aware recommender system could be effectively designed. Past studies seem to have failed to survey on enough attention to the role of context in designing recommender systems for digital libraries.Index Terms-Recommender system, digital library, context, information seeking in context.
I. INTRODUCTIONToday, access to useful and relevant information is a prevalent problem due to information overload [1]. Recommender Systems (RSs) are a special class of personalized systems that aim at predicting a user's interest in available products and services by relying on information seeking behavior and previously rated items or item features [1] . On the whole, they are mainly intelligent applications, created to support users by personalized recommendations in search process and their decision-making while interacting with large information spaces.RSs are used in digital libraries, electronic stores, travel tours, restaurants, hospitals and in general can be useful in any decision-making process to provide predictions of appropriate items to specific users [2]. Moreover, dramatic increase in unorganized information on the web has made such systems a prerequisite for the information seeking on the World Wide Web. A study of 96 RSs by Rao and Talwa [3] showed that RSs have been developed in various domains including web recommendation application, movie/TV, information/document recommendation, Usenet news recommendation, information filtering and sharing, music recommendation, restaurant recommendation application, organizational expertise recommendation, personalized newspaper domain, e-commerce application, travel recommendation, electronic catalogue item recommendation, learning resources recommendation, Web search and filtering, virtual fair recommendation, jokes recommendation, and image RSs. Also a taxonomy and comprehensive explanation of elements in recommender agents on the Internet has been provided by Montaner, Lopez and De la Rosa [4].During a commercial interaction, RSs have advantages for both customers and merchants. For example, in a business interaction through the online shopping, RSs recommend customers items so help customers to find their favorite items among an overwhelming number of items in an electronic department store; therefore RSs can facilitate and accelerate shopping for users. Also, merchant...