{"id":"W4382994301","doi":"10.2139/ssrn.4498130","title":"Pseudo Oversampling Based on Feature Transformation and Fuzzy Membership Functions for Imbalanced and Overlapping Data","year":2023,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Imbalanced Data Classification Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"","keywords":"Oversampling; Feature (linguistics); Transformation (genetics); Fuzzy logic; Artificial intelligence; Pattern recognition (psychology); Computer science; Data mining; Mathematics; Biology; Linguistics","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":["metaepi_narrow","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.0025916,0.000307283,0.0003114385,0.0003620968,0.0004330619,0.0005983465,0.001267569,0.0002943622,9.817552e-7],"category_scores_gemma":[0.0001908116,0.0003017494,0.00007863141,0.0002354857,0.00004631787,0.0008480807,0.0004085597,0.002655188,0.000002521706],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006033581,"about_ca_system_score_gemma":0.001413763,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001963729,"about_ca_topic_score_gemma":0.0001261221,"domain_scores_codex":[0.997184,0.00009174462,0.0003667912,0.0007997131,0.0003632503,0.00119451],"domain_scores_gemma":[0.997952,0.0002805073,0.0003897189,0.001148069,0.0001291651,0.0001005613],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003972974,0.0001687059,0.0007406081,0.0009020089,0.0006430555,0.00000519031,0.001193355,0.003509288,0.00265817,0.8285367,0.009330239,0.1519154],"study_design_scores_gemma":[0.001378598,0.0003280211,0.001598307,0.0004986053,0.0001078,0.0001384526,0.0004969646,0.5491365,0.0001392718,0.4415551,0.003933306,0.0006891504],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0009206728,0.0003896787,0.9894304,0.007342164,0.0004603522,0.000633673,0.0002832056,0.0003765494,0.0001632956],"genre_scores_gemma":[0.8719953,0.007936474,0.1147892,0.000781985,0.0007992852,0.0002457919,0.00254172,0.0001288082,0.0007813695],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8746412,"threshold_uncertainty_score":0.9999434,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05631966215033283,"score_gpt":0.3033271105729997,"score_spread":0.2470074484226668,"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."}}