Classification of multi-view network data using quantum walk neural network

Abstract

With the advent of information stored in the form of network structured data, large-scale graph based machine learning techniques have gained sudden relevance in the field of artificial intelligence. These techniques have been used to solve problems from very diverse domains, such as humanitarian response, poverty estimation, urban planning, epidemic containment, etc. Yet the vast majority of computational tools and algorithms used in these applications do not account for the multi-view nature of social networks: people are related in myriad ways, but most graph learning models treat relations as binary. In this work, we propose a graph-based Quantum walk inspired neural network for learning on multi-view networks by introducing preservation and collaboration parameters in subspace merging and effectively optimizing them using Non Dominated Sorting Genetic Algorithm. We show that this method outperforms state-of-the-art learning algorithms on multi-view networks.