{"id":"W4375869758","doi":"10.48550/arxiv.1907.01136","title":"Finding Outliers in Gaussian Model-Based Clustering","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Advanced Clustering Algorithms Research","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McMaster University","funders":"","keywords":"Outlier; Cluster analysis; Gaussian; Computer science; Mixture model; Pattern recognition (psychology); Artificial intelligence; Statistics; Econometrics; Mathematics; Data mining; Physics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.007018635,0.001246178,0.002523896,0.00398581,0.001835351,0.002918152,0.003512198,0.002979617,0.0008766217],"category_scores_gemma":[0.0368637,0.001118194,0.001301025,0.004737682,0.002749478,0.002991409,0.003622176,0.003041722,0.0009520982],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002133963,"about_ca_system_score_gemma":0.001913307,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01062044,"about_ca_topic_score_gemma":0.008371744,"domain_scores_codex":[0.9930714,0.002482718,0.0003275656,0.001250189,0.002578511,0.0002895872],"domain_scores_gemma":[0.9891166,0.00572969,0.001059809,0.001682501,0.002104956,0.0003064606],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0002397425,0.00007879166,0.00715805,0.0004182202,0.0002650541,0.0002076874,0.0007883848,0.7065602,0.002189224,0.08123506,0.008992808,0.1918668],"study_design_scores_gemma":[0.000009470974,0.00002350152,0.0006753136,0.00004442686,0.00001659246,0.00007641585,0.0001042513,0.9173184,0.0009078048,0.07819916,0.002594155,0.00003050985],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.006954926,0.0007897139,0.9906792,0.0002387255,0.00004967651,0.00003704171,0.0001002517,0.0006600627,0.0004903359],"genre_scores_gemma":[0.2981112,0.00192963,0.6946115,0.0003649948,0.0002645792,0.0002857783,0.001594755,0.0006096341,0.002227875],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01062044,"threshold_uncertainty_score":0.03711855,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1026567429127034,"score_gpt":0.2399083167034491,"score_spread":0.1372515737907457,"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."}}