BenchTemp: A General Benchmark for Evaluating Temporal Graph Neural Networks

TL;DR

BenchTemp is a comprehensive benchmark framework that standardizes datasets, evaluation pipelines, and workload diversity to enable fair and extensive comparison of temporal graph neural networks across various tasks and settings.

Contribution

It introduces a unified benchmark with standardized datasets and evaluation pipelines for fair comparison of TGNNs across multiple tasks and settings.

Findings

Extensive comparison of TGNN models on various tasks.

Identification of strengths and weaknesses of different TGNNs.

Benchmarking results highlighting efficiency and effectiveness differences.

Abstract

To handle graphs in which features or connectivities are evolving over time, a series of temporal graph neural networks (TGNNs) have been proposed. Despite the success of these TGNNs, the previous TGNN evaluations reveal several limitations regarding four critical issues: 1) inconsistent datasets, 2) inconsistent evaluation pipelines, 3) lacking workload diversity, and 4) lacking efficient comparison. Overall, there lacks an empirical study that puts TGNN models onto the same ground and compares them comprehensively. To this end, we propose BenchTemp, a general benchmark for evaluating TGNN models on various workloads. BenchTemp provides a set of benchmark datasets so that different TGNN models can be fairly compared. Further, BenchTemp engineers a standard pipeline that unifies the TGNN evaluation. With BenchTemp, we extensively compare the representative TGNN models on different tasks (e.g., link prediction and node classification) and settings (transductive and inductive), w.r.t. both effectiveness and efficiency metrics. We have made BenchTemp publicly available at https://github.com/qianghuangwhu/benchtemp.