Buy custom Using Memory essay
Using Memory to Reduce the Information Overload in a University Digital Library
Information overload problem is the difficulty of gathering relevant information from the huge amount of information coming in. University Digital Libraries (UDL) let users develop an interest profile and the UDL alerts the users about information based on their interest as it is added to the library. However, since the information volume is large, users often end up receiving irrelevant items.
A recommender system evaluates and filters all available information according to the preferences set by the user in his profile. Two main classifications of recommender systems are:
- Content-based – Recommends information based on content and user’s past experience dealing with similar items.
- Collaborative – Information is recommended based on the user’s social environment, ignoring the content, and is based on other user’s recommendations with similar profiles. This filter allows users to share their experience and rate information.
Though recommender systems have made the process of selection and short-listing of relevant information easier, the problem still remains. This article suggests the implementation of an improved recommender system to overcome information overload in a UDL. This system, a hybrid of content based and collaborative systems, also presents a memory to remember selected but previously not recommended items, and incorporates them in future recommendations. The users are asked for their feedback on recommendations, based on which their profiles are automatically updated for future references. This recommender system takes into consideration the following aspects: importance degree of discipline, relevance degree of resource, compatibility degree between two users, preference degree of one source over another and qualitative number of resources that the user chooses to receive. (Tejeda-Lorente, Porcel, Martínez, López-Herrera, & Herrera-Viedma, 2011)