Embedding-Based Deep Neural Network and Convolutional Neural Network Graph Classifiers

One of the most significant graph data analysis tasks is classification, as graphs are complex structures used for illustrating relationships between entity pairs. Graphs essential in many domains, such description chemical molecules, biological networks, social relationships, etc. Real-world complicated and large. As a result, there need to find way represent or encode graph’s structure so that it can be easily utilized by machine learning models. Therefore, embedding considered one powerful solutions representation. Inspired Doc2Vec model Natural Language Processing (NLP), this paper first investigates different ways (sub)graph each subgraph fixed-length feature vector, which then input any classifier. Thus, two supervised classifiers—a deep neural network (DNN) convolutional (CNN)—are proposed enhance classification. Experimental results on five benchmark datasets indicate models obtain competitive superior some traditional classification methods deep-learning-based approaches three out datasets, with an impressive accuracy rate 94% NCI1 dataset.