Network Embeddings for Graph Classification Problem

A wide range of real world applications deal with network analysis and classification tasks. An ease of representing data with graphs makes them very valuable asset in any data mining toolbox; however, the complexity of working with graphs led researchers to seek for new ways of representing and analyzing graphs, of which network embeddings have become broadly popular due to their success in several machine learning areas such as graph classification, visualization, and pattern recognition. In this talk, we will present a history of solutions to graph classification problem, starting from graph kernels to the latest distributed embeddings based on convolutional neural networks and auto-encoders. In addition, we will present our recent result on anonymous walk embeddings, which show substantial uplift in classification accuracy to state-of-the-art algorithms. We will wrap up with a discussion of network embeddings applications and future directions.