{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005439923,0.001951688,0.002765374,0.003704362,0.002229169,0.002587994,0.004858904,0.00273154,0.001586701],"category_scores_gemma":[0.02433899,0.001283547,0.001509829,0.004384424,0.001123994,0.003502525,0.002725034,0.002286151,0.002112248],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001361152,"about_ca_system_score_gemma":0.001954215,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008434593,"about_ca_topic_score_gemma":0.02007399,"domain_scores_codex":[0.9909356,0.003301781,0.0005063299,0.002864419,0.002019084,0.0003728759],"domain_scores_gemma":[0.9772676,0.01177539,0.001550942,0.005734828,0.00333289,0.0003383562],"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":[0.001014646,0.0004284694,0.01062004,0.001081173,0.001007547,0.0007776323,0.001046323,0.4387459,0.02353351,0.02098193,0.02915513,0.4716078],"study_design_scores_gemma":[0.00001663796,0.00002748507,0.001291234,0.00003593782,0.00006468026,0.0001602277,0.00009981388,0.9639262,0.004560468,0.02706345,0.002731909,0.00002198495],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02597542,0.0004308776,0.9697579,0.0002288131,0.0001120411,0.0000963786,0.0006489176,0.001450334,0.001299365],"genre_scores_gemma":[0.4628237,0.0003684089,0.5230433,0.0002989269,0.0002576637,0.0003677786,0.005789774,0.0007989487,0.006251461],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.008434593,"threshold_uncertainty_score":0.02876937,"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."}}