{ "@graph" : [ { "@id" : "http://semantics.gr/authorities/EKT-voc-classifier/605963148", "@type" : "skos:Concept", "broader" : "http://semantics.gr/authorities/EKT-voc-classifier/1532468312", "closeMatch" : "http://id.loc.gov/authorities/subjects/sh94004659", "exactMatch" : [ "http://id.loc.gov/authorities/subjects/sh85008180", "http://vocabularies.unesco.org/thesaurus/concept3052", "http://semantics.gr/authorities/EKT-voc/605963148" ], "note" : { "@language" : "en", "@value" : "isi - Computer Science, Artificial Intelligence covers resources that focus on research and techniques to create machines that attempt to efficiently reason, problem-solve, use knowledge representation, and perform analysis of contradictory or ambiguous information. This category includes resources on artificial intelligence technologies such as expert systems, fuzzy systems, natural language processing, speech recognition, pattern recognition, computer vision, decision-support systems, knowledge bases, and neural networks." }, "prefLabel" : [ { "@language" : "en", "@value" : "Artificial Intelligence" }, { "@language" : "el", "@value" : "Τεχνητή νοημοσύνη" } ], "relatedMatch" : [ "http://id.loc.gov/authorities/subjects/sh85031234", "http://id.loc.gov/authorities/subjects/sh85079324" ] }, { "@id" : "http://semantics.gr/authorities/openarchives-item-types/metaptyxiakh-ergasia", "@type" : "skos:Concept", "broader" : "http://semantics.gr/authorities/openarchives-item-types/Research-Paper-", "exactMatch" : "http://vocab.getty.edu/aat/300077723", "prefLabel" : [ { "@language" : "en", "@value" : "Master thesis" }, { "@language" : "el", "@value" : "Μεταπτυχιακή εργασία" } ] }, { "@id" : "https://www.openarchives.gr/aggregator-openarchives/edm/aggregation/provider/000004-2159_30975%231", "@type" : "ore:Aggregation", "aggregatedCHO" : "https://www.openarchives.gr/aggregator-openarchives/edm/psepheda/000004-2159_30975", "dataProvider" : "Πανεπιστήμιο Μακεδονίας", "isShownAt" : "https://dspace.lib.uom.gr/handle/2159/30975", "provider" : "Greek Aggregator OpenArchives.gr | National Documentation Centre (EKT)", "rights" : "http://creativecommons.org/licenses/by-nc-sa/4.0/" }, { "@id" : "https://www.openarchives.gr/aggregator-openarchives/edm/psepheda/000004-2159_30975", "@type" : "edm:ProvidedCHO", "contributor" : { "@language" : "el", "@value" : "Χρήστου-Βαρσακέλης, Δημήτριος" }, "creator" : { "@language" : "el", "@value" : "Καπεταδημήτρη, Γεωργία" }, "description" : [ { "@language" : "en", "@value" : "Submitted by ΓΕΩΡΓΙΑ ΚΑΠΕΤΑΔΗΜΗΤΡΗ (aid23007@uom.edu.gr) on 2024-07-19T08:03:39Z\nNo. of bitstreams: 2\nlicense_rdf: 1031 bytes, checksum: 934f4ca17e109e0a05eaeaba504d7ce4 (MD5)\nKapetadimitriGeorgiaMsc2024.pdf: 6339334 bytes, checksum: 68ed2894f28e7c662386d6ed8964532b (MD5)" }, { "@language" : "en", "@value" : "Made available in DSpace on 2024-07-19T08:19:51Z (GMT). No. of bitstreams: 2\nlicense_rdf: 1031 bytes, checksum: 934f4ca17e109e0a05eaeaba504d7ce4 (MD5)\nKapetadimitriGeorgiaMsc2024.pdf: 6339334 bytes, checksum: 68ed2894f28e7c662386d6ed8964532b (MD5)\n Previous issue date: 2024-07-19" }, { "@language" : "en", "@value" : "Graph Neural Networks are widely employed for node classification in\n attributed networks. When it comes to fraud detection, however, GNNs\n can perform poorly, because a node’s features are typically computed\n based on its local neighborhood, and this allows fraudsters to \"blend in\"\n among legitimate users. In this thesis, GNNs and supervised contrastive\n learning are proposed for fraud detection on datasets where fraudsters\n mayuse intricate strategies to camouflage themselves within the network.\n We train our GNNs using novel structural features in addition to those\n typically used in similar studies. The proposed features are based on the\n empirical probability distributions of various graph structural attributes\n which are extracted from a given dataset. We also apply supervised\n contrastive learning, enhanced with synthetic samples for the minority\n class (i.e., the fraudsters). Under our approach, the classifying capability of\n the GNN(measured via F1-macro, AUC, Recall) is improved by boosting\n the representation power of the calculated embeddings that maximize\n the similarity between legitimate users while minimizing that between\n fraudsters and legitimate users. Numerical experiments on two real-world\n multi-relation graph datasets (Amazon and YelpChi) demonstrate the\n effectiveness of the proposed method, whose improvements over the\n state-of the-art were especially significant in the larger YelpChi dataset." }, { "@language" : "el", "@value" : "Διπλωματική εργασία--Πανεπιστήμιο Μακεδονίας, Θεσσαλονίκη, 2024." }, { "@language" : "en", "@value" : "Approved for entry into archive by Κυριακή Μπαλτά (balta@uom.gr) on 2024-07-19T08:19:51Z (GMT) No. of bitstreams: 2\nlicense_rdf: 1031 bytes, checksum: 934f4ca17e109e0a05eaeaba504d7ce4 (MD5)\nKapetadimitriGeorgiaMsc2024.pdf: 6339334 bytes, checksum: 68ed2894f28e7c662386d6ed8964532b (MD5)" } ], "identifier" : "http://dspace.lib.uom.gr/handle/2159/30975", "publisher" : { "@language" : "el", "@value" : "Πανεπιστήμιο Μακεδονίας" }, "dc:rights" : [ { "@language" : "en", "@value" : "http://creativecommons.org/licenses/by-nc-sa/4.0/" }, { "@language" : "el", "@value" : "Αναφορά Δημιουργού - Μη Εμπορική Χρήση - Παρόμοια Διανομή 4.0 Διεθνές" } ], "subject" : [ { "@language" : "el", "@value" : "Contrastive Learning" }, { "@id" : "http://semantics.gr/authorities/EKT-voc-classifier/605963148" }, { "@language" : "el", "@value" : "Graph Neural Networks" } ], "title" : { "@language" : "en", "@value" : "Enhancing fraud detection via gnns with synthetic fraud node generation and integrated structural features" }, "type" : "http://semantics.gr/authorities/openarchives-item-types/metaptyxiakh-ergasia", "dc:type" : [ { "@language" : "en", "@value" : "Electronic Thesis or Dissertation" }, { "@language" : "en", "@value" : "Text" } ], "created" : "2024-07-19T08:19:51Z" } ], "@context" : { "rights" : { "@id" : "http://www.europeana.eu/schemas/edm/rights", "@type" : "@id" }, "provider" : { "@id" : "http://www.europeana.eu/schemas/edm/provider" }, "isShownAt" : { "@id" : "http://www.europeana.eu/schemas/edm/isShownAt", "@type" : "@id" }, "dataProvider" : { "@id" : "http://www.europeana.eu/schemas/edm/dataProvider" }, "aggregatedCHO" : { "@id" : "http://www.europeana.eu/schemas/edm/aggregatedCHO", "@type" : "@id" }, "contributor" : { "@id" : "http://purl.org/dc/elements/1.1/contributor" }, "type" : { "@id" : "http://purl.org/dc/elements/1.1/type", "@type" : "@id" }, "description" : { "@id" : "http://purl.org/dc/elements/1.1/description" }, "subject" : { "@id" : "http://purl.org/dc/elements/1.1/subject" }, "title" : { "@id" : "http://purl.org/dc/elements/1.1/title" }, "publisher" : { "@id" : "http://purl.org/dc/elements/1.1/publisher" }, "creator" : { "@id" : "http://purl.org/dc/elements/1.1/creator" }, "created" : { "@id" : "http://purl.org/dc/terms/created" }, "identifier" : { "@id" : "http://purl.org/dc/elements/1.1/identifier" }, "exactMatch" : { "@id" : "http://www.w3.org/2004/02/skos/core#exactMatch", "@type" : "@id" }, "broader" : { "@id" : "http://www.w3.org/2004/02/skos/core#broader", "@type" : "@id" }, "prefLabel" : { "@id" : "http://www.w3.org/2004/02/skos/core#prefLabel" }, "relatedMatch" : { "@id" : "http://www.w3.org/2004/02/skos/core#relatedMatch", "@type" : "@id" }, "note" : { "@id" : "http://www.w3.org/2004/02/skos/core#note" }, "closeMatch" : { "@id" : "http://www.w3.org/2004/02/skos/core#closeMatch", "@type" : "@id" }, "ore" : "http://www.openarchives.org/ore/terms/", "owl" : "http://www.w3.org/2002/07/owl#", "svcs" : "http://rdfs.org/sioc/services#", "skos" : "http://www.w3.org/2004/02/skos/core#", "rdfs" : "http://www.w3.org/2000/01/rdf-schema#", "rdaGr2" : "http://rdvocab.info/ElementsGr2/", "edm" : "http://www.europeana.eu/schemas/edm/", "ekt" : "https://www.semantics.gr/authorities/schemanamespaces/ekt#", "doap" : "http://usefulinc.com/ns/doap#", "rdf" : "http://www.w3.org/1999/02/22-rdf-syntax-ns#", "xalan" : "http://xml.apache.org/xalan", "dcterms" : "http://purl.org/dc/terms/", "wgs84_pos" : "http://www.w3.org/2003/01/geo/wgs84_pos#", "foaf" : "http://xmlns.com/foaf/0.1/", "crm" : "http://www.cidoc-crm.org/rdfs/cidoc_crm_v5.0.2_english_label.rdfs#", "dc" : "http://purl.org/dc/elements/1.1/" } }