{"id":"W4395018386","doi":"10.1109/icosst60641.2023.10505068","title":"Overcoming Imbalanced Datasets and Feature Complexity in Fraud Transaction Detection Through Down-Sampling and Dimension Reduction","year":2023,"lang":"en","type":"article","venue":"","topic":"Imbalanced Data Classification Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Bishop's University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Database transaction; Reduction (mathematics); Dimension (graph theory); Dimensionality reduction; Feature (linguistics); Data mining; Sampling (signal processing); Pattern recognition (psychology); Artificial intelligence; Mathematics; Database; Detector","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.006015555,0.001066377,0.001881639,0.00267613,0.0007839507,0.002616064,0.001399485,0.0008832028,0.000423849],"category_scores_gemma":[0.02315085,0.0004652993,0.001093295,0.002161208,0.0009516635,0.002673514,0.002248038,0.001983091,0.0002910508],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007942296,"about_ca_system_score_gemma":0.001196362,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00237711,"about_ca_topic_score_gemma":0.00234824,"domain_scores_codex":[0.9962812,0.001626025,0.000307948,0.0004953743,0.001016704,0.0002726679],"domain_scores_gemma":[0.9885677,0.005410839,0.00100838,0.002955572,0.001797241,0.0002601577],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000590768,0.0008630751,0.04013571,0.0002705208,0.0003385729,0.0004330124,0.0008394521,0.4306017,0.0144166,0.02102624,0.004418961,0.4860654],"study_design_scores_gemma":[0.00001073446,0.0000932296,0.002860453,0.00002201344,0.00003446806,0.00009509351,0.00009121484,0.9824133,0.002519805,0.01102235,0.0008172477,0.0000201248],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1556494,0.0008654117,0.8404246,0.0009972566,0.0001135974,0.0002147999,0.0002456032,0.0004737689,0.001015596],"genre_scores_gemma":[0.7477266,0.0005247176,0.2498358,0.0002239549,0.0002120621,0.0001823249,0.0006873092,0.00004467609,0.0005625612],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006015555,"threshold_uncertainty_score":0.03181368,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06048644076122597,"score_gpt":0.3133627878569141,"score_spread":0.2528763470956881,"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."}}