Neighbor2vec: an efficient and effective method for Graph Embedding
This work addresses the scalability and effectiveness limitations of existing unsupervised graph embedding techniques for tasks like node classification and link prediction, though it appears incremental as it builds on neighbor-based sampling strategies.
The paper tackles the problem of capturing network patterns in graph embedding by proposing neighbor2vec, a neighbor-based sampling strategy that learns node neighborhood representations through feature propagation, resulting in average accuracy improvements of up to 6.8% in node classification and 3.0% in link prediction over competing methods.
Graph embedding techniques have led to significant progress in recent years. However, present techniques are not effective enough to capture the patterns of networks. This paper propose neighbor2vec, a neighbor-based sampling strategy used algorithm to learn the neighborhood representations of node, a framework to gather the structure information by feature propagation between the node and its neighbors. We claim that neighbor2vec is a simple and effective approach to enhancing the scalability as well as equality of graph embedding, and it breaks the limits of the existing state-of-the-art unsupervised techniques. We conduct experiments on several node classification and link prediction tasks for networks such as ogbn-arxiv, ogbn-products, ogbn-proteins, ogbl-ppa,ogbl-collab and ogbl-citation2. The result shows that Neighbor2vec's representations provide an average accuracy scores up to 6.8 percent higher than competing methods in node classification tasks and 3.0 percent higher in link prediction tasks. The neighbor2vec's representations are able to outperform all baseline methods and two classical GNN models in all six experiments.