{"id":"W3001278818","doi":"10.1109/dsaa.2019.00034","title":"A Study on the Impact of Data Characteristics in Imbalanced Regression Tasks","year":2019,"lang":"en","type":"article","venue":"","topic":"Imbalanced Data Classification Techniques","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"","keywords":"Computer science; Regression; Regression analysis; Artificial intelligence; Machine learning; Statistics; Mathematics","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":[],"consensus_categories":[],"category_scores_codex":[0.0006600553,0.0001044024,0.0001835555,0.00009083198,0.00002035492,0.00005152499,0.002645855,0.00003351589,0.00003729399],"category_scores_gemma":[0.0001054819,0.00005380785,0.00002274206,0.000363281,0.0000201104,0.0004704106,0.0007217013,0.0001419346,0.00006041773],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004399407,"about_ca_system_score_gemma":0.00006146298,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007356036,"about_ca_topic_score_gemma":0.000003414174,"domain_scores_codex":[0.9988651,0.0001170529,0.0002764382,0.0003653806,0.0002335257,0.0001425366],"domain_scores_gemma":[0.9962068,0.0001724832,0.0001852691,0.003367407,0.00004550246,0.00002257713],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0001080419,0.002338648,0.8379347,0.00001936864,0.00005561737,0.00001694741,0.002016796,0.000006114078,0.04613221,0.04919815,0.01992369,0.04224969],"study_design_scores_gemma":[0.0003586834,0.0003670368,0.9771332,0.00005009332,0.000001168739,0.000001539934,0.0001094139,0.01824763,0.003129293,0.0004160408,0.0000802278,0.0001057029],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9748356,0.00000335223,0.02133893,0.0004122069,0.00006846624,0.0007866629,0.00006300845,0.0001282876,0.002363522],"genre_scores_gemma":[0.9967402,0.000005143716,0.003013451,0.00006669114,0.000006553213,0.00001417475,0.00003390589,0.000005165121,0.0001146887],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1391984,"threshold_uncertainty_score":0.4916701,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08241202041512183,"score_gpt":0.3719947321635921,"score_spread":0.2895827117484703,"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."}}