Papers › MGTAB: A Multi-Relational Graph-Based Twitter Account Detection Benchmark

MGTAB: A Multi-Relational Graph-Based Twitter Account Detection Benchmark

3 Jan 2023arXiv:2301.01123archive 2025-07-28

Shuhao Shi, Kai Qiao, Jian Chen, Shuai Yang, Jie Yang, Baojie Song, Linyuan Wang, Bin Yan

The development of social media user stance detection and bot detection methods rely heavily on large-scale and high-quality benchmarks. However, in addition to low annotation quality, existing benchmarks generally have incomplete user relationships, suppressing graph-based account detection research. To address these issues, we propose a Multi-Relational Graph-Based Twitter Account Detection Benchmark (MGTAB), the first standardized graph-based benchmark for account detection. To our knowledge, MGTAB was built based on the largest original data in the field, with over 1.55 million users and 130 million tweets. MGTAB contains 10,199 expert-annotated users and 7 types of relationships, ensuring high-quality annotation and diversified relations. In MGTAB, we extracted the 20 user property features with the greatest information gain and user tweet features as the user features. In addition, we performed a thorough evaluation of MGTAB and other public datasets. Our experiments found that graph-based approaches are generally more effective than feature-based approaches and perform better when introducing multiple relations. By analyzing experiment results, we identify effective approaches for account detection and provide potential future research directions in this field. Our benchmark and standardized evaluation procedures are freely available at: https://github.com/GraphDetec/MGTAB.

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

graphdetec/mgtab officialmentioned in paperpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Node ClassificationStance DetectionTwitter Bot Detection

Datasets

Introduced by this paper, per the archive.

MGTAB

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Stance Detection MGTAB RGT Acc 87.8 #1 of 4 Archive leaderboard report
Stance Detection MGTAB RGT F1 86.9 #1 of 4 Archive leaderboard report
Stance Detection MGTAB Simple-HGN Acc 85.3 #2 of 4 Archive leaderboard report
Stance Detection MGTAB Simple-HGN F1 84.4 #2 of 4 Archive leaderboard report
Stance Detection MGTAB GCN Acc 82.4 #3 of 4 Archive leaderboard report
Stance Detection MGTAB GCN F1 81.5 #3 of 4 Archive leaderboard report
Stance Detection MGTAB GAT Acc 82.2 #4 of 4 Archive leaderboard report
Stance Detection MGTAB GAT F1 81 #4 of 4 Archive leaderboard report
Twitter Bot Detection MGTAB RGT Acc 92.1 #1 of 4 Archive leaderboard report
Twitter Bot Detection MGTAB RGT F1 90.4 #1 of 4 Archive leaderboard report
Twitter Bot Detection MGTAB BotRGCN Acc 89.6 #2 of 4 Archive leaderboard report
Twitter Bot Detection MGTAB BotRGCN F1 87.2 #2 of 4 Archive leaderboard report
Twitter Bot Detection MGTAB GAT Acc 87 #3 of 4 Archive leaderboard report
Twitter Bot Detection MGTAB GAT F1 82.3 #3 of 4 Archive leaderboard report
Twitter Bot Detection MGTAB GCN Acc 85.8 #4 of 4 Archive leaderboard report
Twitter Bot Detection MGTAB GCN F1 78.3 #4 of 4 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

GATGCNRGCN

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections