2022
DOI: 10.1109/tkde.2020.3025588
|View full text |Cite
|
Sign up to set email alerts
|

Graph Neural Network for Fraud Detection via Spatial-Temporal Attention

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
15
0
2

Year Published

2022
2022
2024
2024

Publication Types

Select...
5
3
1

Relationship

0
9

Authors

Journals

citations
Cited by 78 publications
(17 citation statements)
references
References 39 publications
0
15
0
2
Order By: Relevance
“…The experimental outcomes revealed that the investigated technique performed well as compared to other methods with regard to an accuracy of 98.40%, AUC (Area under receiver operating characteristic curve) of 92.88%, precision of 97.34% and F1-score of 56.95% for detecting the CCF. D. Cheng, et.al (2022) introduced a STAGN (spatial-temporal attention-based graph network) in order to detect fraud in credit card [17]. Generally, a GNN (graph neural network) for learning graph attributes.…”
Section: Literature Reviewmentioning
confidence: 99%
“…The experimental outcomes revealed that the investigated technique performed well as compared to other methods with regard to an accuracy of 98.40%, AUC (Area under receiver operating characteristic curve) of 92.88%, precision of 97.34% and F1-score of 56.95% for detecting the CCF. D. Cheng, et.al (2022) introduced a STAGN (spatial-temporal attention-based graph network) in order to detect fraud in credit card [17]. Generally, a GNN (graph neural network) for learning graph attributes.…”
Section: Literature Reviewmentioning
confidence: 99%
“…Furthermore, the relationship between multiple instances, as transactions or customers, is also embedded into this principle. This principle attends fraud expert tasks when analyzing complex fraud cases involving multiple actors and comparing past patterns, which is constantly performed during their analyses whether through Feature Impact EM or network and graph visualizations of multiple customer transactions over time [16,9,26]. Without DP 5, experts would lack such features for understanding AI with temporal and multiple feature and instances perspectives.…”
Section: Dp3 Dp 3 Instantiation Enables Experts To Look Closely Into What Data Featuresmentioning
confidence: 99%
“…Bu durum ise, çevrim içi dolandırıcılık tespiti sistemindeki dolandırıcılık davranışlarına en yakın dolandırıcılık davranış modellerine odaklanmayı gerektirir. Bu nedenle [9] kredi kartı dolandırıcılık tespiti için mekansal-zamansal dikkat tabanlı bir grafik ağı (STAGN) modeli önerilmiştir. Daha sonra 3 boyutlu bir evrişim ağına beslenen öğrenilmiş temsillerinin üzerine uzamsal-zamansal dikkat kullanılmıştır.…”
Section: Literatür Taramasıunclassified