Clustering of learning objects with Self-Organizing Maps

The increasing availability of digital educational resources in the Internet, called learning objects, has been followed by the definition of indexing standards. However, the lack of consensus about the definition of learning objects, as well the diversity of metadata approaches for its classification hinders the selection process of these elements. This scenario requires new investigations that allow the establishment of parameters for the creation of a specific model of artificial neural network for the clustering of learning objects. The implementation of this model is related to a theoretical-methodological approach, based on standard metadata criteria, which makes the formation of input samples possible for the construction of a Self-Organizing Map (SOM - Kohonen model) through algorithms and mathematical models. Consequently, the development of this proposal for the clustering of learning objects can support the educational work in face-to-face and online environments and collaborate with the reusability of learning objects. Another goal of this research was the determination of a weight mask, one of the Kohonen model's parameters, and how it would affect the final result. For that, a comparison was made between the training results with and without the mask, showing the relevance of this method for obtaining better clustering results.