Folksonomy-Based Indexing for Location-Aware Retrieval of Learning Contents

With the fast development of wireless communication and sensor technologies, ubiquitous learning has become a promising learning paradigm. In context-aware ubiquitous learning environments, it is desirable that learning content is retrieved according to environmental contexts, such as learners' location. However, traditional information retrieval schemes are not designed for content retrieval in ubiquitous learning environments. Recently, folksonomies have emerged as a successful kind of applications for categorizing web resources in a collaborative manner. This paper focuses on the index creation problem for location-aware learning content retrieval. First, we propose a bottom-up approach to constructing the index according to the similarity between tags, which considers metadata and structural information of the teaching materials annotated by the tags. Then, a maintenance mechanism is designed to efficiently update the index. The index creation method has been implemented, and a synthetic learning object repository has been built to evaluate the proposed approach. Experimental results show that this method can increase precision of retrieval. In addition, impacts of different similarity functions on precision are discussed.