Crosswalk between educational metadata standards

The formats diversity used for the development of learning objects contributes to the relevant information retrieval process remains a major challenge for Information Science and Computer Science. The adoption of metadata for cataloging these resources are being used both nationally and internationally. However, there is no consensus regarding the best metadata standard. It aims to present a crosswalk between main metadata standards used by learning objects repositories. Methodology is based on exploratory and analytical analysis of learning object repositories to identify the metadata of each initiative and propose a correlation model between the patterns. The results indicate that Dublin Core and LOM standards are references to other metadata standards. Thus, mapping was performed based on a correspondence between these patterns and validating was analyzed 140 learning objects available in repositories web. It follows that it is possible to establish a correlation model between the metadata with a significant number of elements to collaborate with the automatic indexing process and improve information retrieval.