{"id":"W4390244867","doi":"10.18280/ria.370606","title":"Evaluating Binary Classification Algorithms on Data Lakes Using Machine Learning","year":2023,"lang":"en","type":"article","venue":"Revue d intelligence artificielle","topic":"Data Stream Mining Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"Department of Artificial Intelligence, Korea University; Lviv Polytechnic National University; Fonds National de la Recherche Luxembourg","keywords":"Computer science; Binary number; Binary classification; Machine learning; Artificial intelligence; Binary data; Data mining; Support vector machine; Mathematics; Arithmetic","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.002260862,0.0002033548,0.0001978479,0.0003370803,0.000440104,0.0002523484,0.002556525,0.00008322317,0.00003743819],"category_scores_gemma":[0.0007899685,0.0002075118,0.00004758501,0.001574315,0.00007204244,0.0007427316,0.001472552,0.0003443899,0.0008625455],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000708726,"about_ca_system_score_gemma":0.00006771401,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006725524,"about_ca_topic_score_gemma":0.000006171845,"domain_scores_codex":[0.997486,0.0002031375,0.0004699054,0.0009704945,0.000396916,0.0004735322],"domain_scores_gemma":[0.9968545,0.0004424993,0.0002162738,0.002305314,0.00009721037,0.0000842299],"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":[0.00001463955,0.000182827,0.0008143783,0.00006497157,0.00002721836,0.0000763577,0.00155055,0.105756,0.06644826,0.01351196,0.001943706,0.8096091],"study_design_scores_gemma":[0.00002052786,0.0001966498,0.00007653273,0.0001408404,0.000007650736,0.00001725272,0.0002254082,0.9743429,0.02089562,0.001001605,0.002851818,0.0002232206],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05504074,0.0001247511,0.9399482,0.000931777,0.0004757196,0.0003474674,0.00005213044,0.001989611,0.001089632],"genre_scores_gemma":[0.6967736,0.0001719951,0.300417,0.0001304318,0.0002138198,0.00004038481,0.0007666755,0.00005942064,0.00142668],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8685868,"threshold_uncertainty_score":0.9999154,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3618588023830966,"score_gpt":0.4242256351558663,"score_spread":0.06236683277276961,"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."}}