{"id":"W6950550149","doi":"10.5281/zenodo.6501988","title":"Practical Difficulties and Applications of GNNs on Financial Data for Fraud Detection","year":2022,"lang":"en","type":"article","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Geophysics and Gravity Measurements","field":"Earth and Planetary Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Larus Technologies (Canada)","funders":"","keywords":"Database transaction; Big data; Key (lock); Financial fraud; Transaction data","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["sts","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0004493982,0.00005164045,0.00006540515,0.00006557396,0.001712399,0.0001050371,0.0003509989,0.00001587368,0.001810199],"category_scores_gemma":[0.0002717445,0.00005219799,0.00001483894,0.0001948137,0.00005769381,0.00009247253,0.0002287494,0.0001187548,0.0001592046],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000007562829,"about_ca_system_score_gemma":0.000003788523,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000066034,"about_ca_topic_score_gemma":0.000005906948,"domain_scores_codex":[0.9991866,0.0001088641,0.0001051634,0.0002230845,0.0002515728,0.000124684],"domain_scores_gemma":[0.9994686,0.00003797684,0.00006510276,0.0002612116,0.0001138803,0.00005321302],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0004121177,0.0003485941,0.0001936885,0.0001198149,0.00004882966,0.000002982467,0.0006051079,0.0008827126,0.002270868,0.005548954,0.0820596,0.9075067],"study_design_scores_gemma":[0.0003175177,0.0005357348,0.02304229,0.000003223886,0.00001370864,0.00001389405,0.000282503,0.002495606,0.0001499762,0.0007281303,0.97233,0.00008739113],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7503313,0.0004009922,0.1306787,0.005414168,0.001019136,0.007836432,0.04319501,0.001055773,0.06006849],"genre_scores_gemma":[0.9960285,0.00001095276,0.0002497483,0.00005245665,0.00007390738,7.712895e-8,0.003445461,0.00005361793,0.00008531129],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9074193,"threshold_uncertainty_score":0.9995872,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06819232942788313,"score_gpt":0.2586478427075545,"score_spread":0.1904555132796714,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}