E2EG: End-to-End Node Classification Using Graph Topology and Text-based Node Attributes

9 Aug 2022  ·  Tu Anh Dinh, Jeroen den Boef, Joran Cornelisse, Paul Groth ·

Node classification utilizing text-based node attributes has many real-world applications, ranging from prediction of paper topics in academic citation graphs to classification of user characteristics in social media networks. State-of-the-art node classification frameworks, such as GIANT, use a two-stage pipeline: first embedding the text attributes of graph nodes then feeding the resulting embeddings into a node classification model. In this paper, we eliminate these two stages and develop an end-to-end node classification model that builds upon GIANT, called End-to-End-GIANT (E2EG). The tandem utilization of a main and an auxiliary classification objectives in our approach results in a more robust model, enabling the BERT backbone to be switched out for a distilled encoder with a 25% - 40% reduction in the number of parameters. Moreover, the model's end-to-end nature increases ease of use, as it avoids the need of chaining multiple models for node classification. Compared to a GIANT+MLP baseline on the ogbn-arxiv and ogbn-products datasets, E2EG obtains slightly better accuracy in the transductive setting (+0.5%), while reducing model training time by up to 40%. Our model is also applicable in the inductive setting, outperforming GIANT+MLP by up to +2.23%.

PDF Abstract

Datasets


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Node Property Prediction ogbn-arxiv E2EG (use raw text) Test Accuracy 0.7362 ± 0.0014 # 41
Validation Accuracy 0.7487 ± 0.0011 # 37
Number of params 83724841 # 6
Ext. data Yes # 1
Node Property Prediction ogbn-products E2EG (use raw text) Test Accuracy 0.8098 ± 0.0040 # 40
Validation Accuracy 0.9234 ± 0.0009 # 36
Number of params 66793520 # 7
Ext. data Yes # 1

Methods