Graph Embedding: A multidimensional mapping of structural relations between entities

The embedding is one of the most innovate discovery in the Natural Language Processing (NLP) field. This technique allows to represent object in a multidimensional Euclidean space. As result, similar object are placed close together.

In the last years, the embedding has been extended also in other domains different from NLP. A successful application is related the social network analysis based on graph theory. In this paper the authors describe a new method to obtain a multidimensional representation of graphs.

In Humanativa we use graph embedding techniques in order to improve our social network analysis algorithms by analyzing structural and function relationship between users.