{"id":"W3144969346","doi":"10.1016/j.ins.2022.05.035","title":"Semi-supervised clustering with inaccurate pairwise annotations","year":2022,"lang":"en","type":"article","venue":"Information Sciences","topic":"Complex Network Analysis Techniques","field":"Physics and Astronomy","cited_by":8,"is_retracted":false,"has_abstract":false,"ca_institutions":"Polytechnique Montréal","funders":"Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro; Conselho Nacional de Desenvolvimento Científico e Tecnológico; Coordenação de Aperfeiçoamento de Pessoal de Nível Superior","keywords":"Pairwise comparison; Computer science; Cluster analysis; Class (philosophy); Relational database; Data mining; Stochastic block model; Link (geometry); Block (permutation group theory); Machine learning; Artificial intelligence; 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":[0.0003133917,0.00006206497,0.00006987478,0.0001489216,0.000771078,0.0001667144,0.0002341802,0.00000404199,0.000802968],"category_scores_gemma":[0.00000234401,0.00005165542,0.00002996906,0.0008473981,0.00006468016,0.00121213,0.0001171621,0.000072488,0.00001696816],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002258284,"about_ca_system_score_gemma":0.00006691596,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001767682,"about_ca_topic_score_gemma":0.00001175909,"domain_scores_codex":[0.9992776,0.00003065617,0.0001915163,0.00007646244,0.0002941347,0.0001296299],"domain_scores_gemma":[0.9996505,0.00003238091,0.0001250822,0.0001078344,0.00005647128,0.00002768303],"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":[0.00003197639,0.00009089614,0.08833807,0.00001396835,0.00006846123,7.666782e-7,0.01142839,0.6513978,0.0001550039,0.03849959,0.01198017,0.1979949],"study_design_scores_gemma":[0.0003106471,0.00015029,0.003345843,0.000009810793,0.00001603096,0.000003008236,0.01213281,0.9229903,0.0002070298,0.002996016,0.05757979,0.0002583915],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5824229,0.00001157207,0.3263107,0.0008286462,0.00009297092,0.0003424842,0.00004888224,0.0002255927,0.08971626],"genre_scores_gemma":[0.9974495,3.069871e-7,0.002154291,0.0001814561,0.00002400675,0.00008972729,0.0000415201,0.000001801994,0.00005744466],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4150266,"threshold_uncertainty_score":0.8791937,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0199553698487683,"score_gpt":0.2664773533025454,"score_spread":0.2465219834537771,"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."}}