Semantic preserving embeddings for multi-relational graphs

In this paper a new machine learning approach to the study of Multi-Relational Graphs as semantic data structures is presented. It shows how vector representations that maintain semantic and topological features of the original data can be obtained from neural encoding architectures and considering the topological properties of the graph. Also, semantic features of these new representations are tested by using some machine learning tasks and new directions on efficient link discovery methodologies on large relational datasets are investigated.