{"id":"W4389322561","doi":"10.21203/rs.3.rs-3688727/v1","title":"OSK: Optimal Subsampling Method Based on K-means Clustering for Imbalanced Big Data","year":2023,"lang":"en","type":"preprint","venue":"Research Square","topic":"Imbalanced Data Classification Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"","keywords":"Cluster analysis; Computer science; Categorical variable; Artificial intelligence; Class (philosophy); Big data; Pattern recognition (psychology); Data mining; Machine learning","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","scholarly_communication","open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.01002711,0.000444249,0.0005755062,0.001252205,0.0004933333,0.001166783,0.01031782,0.0004740348,0.000008074716],"category_scores_gemma":[0.003112125,0.000452564,0.0001783644,0.001212008,0.0001196056,0.0003939419,0.01186904,0.001924035,0.00009245146],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004916224,"about_ca_system_score_gemma":0.0009909186,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001630293,"about_ca_topic_score_gemma":0.00006160067,"domain_scores_codex":[0.9925967,0.0008376511,0.0006607416,0.002766186,0.001855874,0.001282889],"domain_scores_gemma":[0.9859263,0.003572492,0.0002652491,0.009068296,0.0009066999,0.0002609161],"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":[0.000574888,0.0007971134,0.0006906514,0.008700204,0.0002632715,0.0001482554,0.00103652,0.3674871,0.005540985,0.01968726,0.1387413,0.4563325],"study_design_scores_gemma":[0.0003704256,0.0001825707,0.0006754538,0.0009652734,0.000006536748,0.00000157298,0.00004463824,0.974193,0.002361247,0.002700147,0.0180571,0.0004420239],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.00004073889,0.00003246092,0.9880055,0.003896675,0.0008598609,0.00256431,0.002530457,0.001753353,0.0003166529],"genre_scores_gemma":[0.01289833,0.0000813007,0.9792166,0.0001350576,0.0006917535,0.001855531,0.004439249,0.0001366175,0.0005455691],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.606706,"threshold_uncertainty_score":0.9998701,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4267506895785027,"score_gpt":0.4980191244722518,"score_spread":0.07126843489374907,"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."}}