Διπλωματική εργασία--Πανεπιστήμιο Μακεδονίας, Θεσσαλονίκη, 2024.
(EL)
Made available in DSpace on 2024-07-19T08:19:51Z (GMT). No. of bitstreams: 2
license_rdf: 1031 bytes, checksum: 934f4ca17e109e0a05eaeaba504d7ce4 (MD5)
KapetadimitriGeorgiaMsc2024.pdf: 6339334 bytes, checksum: 68ed2894f28e7c662386d6ed8964532b (MD5)
Previous issue date: 2024-07-19
(EN)
Approved for entry into archive by Κυριακή Μπαλτά (
[email protected]) on 2024-07-19T08:19:51Z (GMT) No. of bitstreams: 2
license_rdf: 1031 bytes, checksum: 934f4ca17e109e0a05eaeaba504d7ce4 (MD5)
KapetadimitriGeorgiaMsc2024.pdf: 6339334 bytes, checksum: 68ed2894f28e7c662386d6ed8964532b (MD5)
(EN)
Submitted by ΓΕΩΡΓΙΑ ΚΑΠΕΤΑΔΗΜΗΤΡΗ (
[email protected]) on 2024-07-19T08:03:39Z
No. of bitstreams: 2
license_rdf: 1031 bytes, checksum: 934f4ca17e109e0a05eaeaba504d7ce4 (MD5)
KapetadimitriGeorgiaMsc2024.pdf: 6339334 bytes, checksum: 68ed2894f28e7c662386d6ed8964532b (MD5)
(EN)
Graph Neural Networks are widely employed for node classification in
attributed networks. When it comes to fraud detection, however, GNNs
can perform poorly, because a node’s features are typically computed
based on its local neighborhood, and this allows fraudsters to "blend in"
among legitimate users. In this thesis, GNNs and supervised contrastive
learning are proposed for fraud detection on datasets where fraudsters
mayuse intricate strategies to camouflage themselves within the network.
We train our GNNs using novel structural features in addition to those
typically used in similar studies. The proposed features are based on the
empirical probability distributions of various graph structural attributes
which are extracted from a given dataset. We also apply supervised
contrastive learning, enhanced with synthetic samples for the minority
class (i.e., the fraudsters). Under our approach, the classifying capability of
the GNN(measured via F1-macro, AUC, Recall) is improved by boosting
the representation power of the calculated embeddings that maximize
the similarity between legitimate users while minimizing that between
fraudsters and legitimate users. Numerical experiments on two real-world
multi-relation graph datasets (Amazon and YelpChi) demonstrate the
effectiveness of the proposed method, whose improvements over the
state-of the-art were especially significant in the larger YelpChi dataset.
(EN)
Πανεπιστήμιο Μακεδονίας
