Fraud Detection with Graph Neural Networks

▬▬ Papers ▬▬▬▬▬▬▬▬▬▬▬▬▬
GNN Survey: arxiv.org/pdf/2106.07178.pdf
DOMINANT: epubs.siam.org/doi/abs/10.113...
StrGNN: arxiv.org/pdf/2005.07427.pdf
xFraud: arxiv.org/pdf/2011.12193.pdf
Github Paper List: github.com/safe-graph/graph-f...
▬▬ Referenced Videos ▬▬▬▬▬▬▬▬
Spatio-Temporal GNNs: • Friendly Introduction ...
Explainable AI on Graphs: • How to explain Graph N...
Graph Transformer: • GNN Project #3.2 - Gra...
▬▬ Used Music ▬▬▬▬▬▬▬▬▬▬▬
Music from Uppbeat (free for Creators!):
uppbeat.io/t/yeti-music/myste...
License code: JDSIDP1BM8GZGN67
▬▬ Used Icons ▬▬▬▬▬▬▬▬▬▬▬
Alle icons are from www.flaticon.com/
by the producers Freepik, Smashicons, Becris, Eucalyp, surang, Flat Icons.
▬▬ Timestamps ▬▬▬▬▬▬▬▬▬▬▬
00:00 Introduction
00:30 Typical Frauds in Networks
01:18 Survey on graph-based fraud detection
01:31 Traditional approaches vs GNN approaches
02:57 Camouflaged frauds
04:02 Building a fraud detection GNN
06:25 Fraud detection GNN papers
10:54 Please vote for next video
▬▬ Support me if you like 🌟
►Link to this channel: bit.ly/3zEqL1W
►Support me on Patreon: bit.ly/2Wed242
►Buy me a coffee on Ko-Fi: bit.ly/3kJYEdl
►E-Mail: [e-mail]
▬▬ My equipment 💻
- Microphone: amzn.to/3DVqB8H
- Microphone mount: amzn.to/3BWUcOJ
- Monitors: amzn.to/3G2Jjgr
- Monitor mount: amzn.to/3AWGIAY
- Height-adjustable table: amzn.to/3aUysXC
- Ergonomic chair: amzn.to/3phQg7r
- PC case: amzn.to/3jdlI2Y
- GPU: amzn.to/3AWyzwy
- Keyboard: amzn.to/2XskWHP
- Bluelight filter glasses: amzn.to/3pj0fK2

[@AnyaSheven]

This is a great overview of fraud detection methods and excellent explanation. Definitely subscribing, thank you!

[@chientruong926]

I’m always impressed with your explainable ability. Thank you a lot and have a nice weekend!!!

[@DeepFindr]

Thanks a lot for your kind words :) have a nice weekend as well!

[@samirelzein1095]

what a pro! Feels like you are a great chef and NNs are your ingredients! ...yum!

[@danielalvey5710]

Fantastic video. Really appreciate the paper links as well.

[@prabhacar]

Very nicely explained. thanks!

[@armandogomis7732]

Just what I'm working on. I've been struggling with the elliptic dataset and the ethreum dataset for months. Thanks for your work, it's great for me.

[@DeepFindr]

Hi! What exactly are you struggling with? Is there something that I may help with? :)

[@armandogomis7732]

@@DeepFindr I have sent you an email. I am a junior data scientist and it's hard for me hahaha. Thanks.

[@isoaso344]

Just out of curiosity, have you ever tried to combine GCNs with Random Forests for elliptic dataset? I think it should work quite well :)

[@armandogomis7732]

@@isoaso344 Yes, but this works much better ;) www.nssce.ac.in/pdf/cse/2017%20batch_poster%202.pdf

[@skanderbegvictor6487]

Your channel is a goldmine please don't stop making videos

[@DeepFindr]

Thanks!

[@Jonnyfoka]

Really great explanations and research. Congrats for that! I am now searching to apply fraud Edge-Detection for a heterogeneous, attributed, dynamic graph, where the edge features are of great importance. Do you happen to know any model implementations?

[@DeepFindr]

Thank you! Sorry but I've not come across anything like that so far.

[@mohammadareeb4826]

Clear and easy explanation. Please make a video on recommendation system using Graph Neural Network using dataset like Movielens.

[@DeepFindr]

Thanks! I'll put it on my list. After the next video I will again make a community poll and include recommender systems there :)

[@jiahao2709]

could you make some video about spectral graph theory?

[@Raresiom]

Hi, can you tell me whcih one the Twitter dataset with fake news is?