A New Approach to Embedding Semantic Link Network with Word2Vec Binary Code

Graph-structured data has come into wide use in various fields where graphs are the natural data structure to model networks. Therefore, the comparison between two graphs becomes a research focus. Traditional approaches for graph comparison face the common problem: either increasing the runtime for large graphs or simplifying the representation of graphs which ignores part of their topological information. In this paper, we build the Semantic Link Network (SLN) to represent documents and introduce a new graph kernel to compare their similarity. Where the graph representations are built according to the co-occurrence relations. And then, the semantic link network will be generated by embedding the rich semantic information which is obtained by neural network language model. Finally, a new graph kernel will be introduced and used to compare the similarity between the semantic link network of documents. The effectiveness and efficiency of this method are evaluated by the document classification task on public corpora and empirical results suggest that the proposed method can achieve better performance than the traditional classification approaches.