{"id":"W4416455941","doi":"10.7717/peerj-cs.3309","title":"The Silhouette coefficient and the Davies-Bouldin index are more informative than Dunn index, Calinski-Harabasz index, Shannon entropy, and Gap statistic for unsupervised clustering internal evaluation of two convex clusters","year":2025,"lang":"en","type":"article","venue":"PeerJ Computer Science","topic":"Advanced Clustering Algorithms Research","field":"Computer Science","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Silhouette; Cluster analysis; Rand index; Pattern recognition (psychology); Ground truth; Consensus clustering; Euclidean distance; Correlation clustering","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":[0.007680979,0.00122019,0.001575414,0.008020844,0.001386555,0.004436408,0.001007914,0.001345059,0.001726517],"category_scores_gemma":[0.05443402,0.0003239923,0.0008849786,0.007005626,0.002714682,0.005257447,0.001803597,0.001422597,0.0006309962],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002239652,"about_ca_system_score_gemma":0.002162755,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004361982,"about_ca_topic_score_gemma":0.00450475,"domain_scores_codex":[0.9916921,0.002309239,0.0008338066,0.001382007,0.00340963,0.000373264],"domain_scores_gemma":[0.9526408,0.02879458,0.005990224,0.003359449,0.008073946,0.001141013],"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":[0.001548167,0.0002883534,0.2418812,0.002071995,0.001237614,0.0005576464,0.002886455,0.1427676,0.01440589,0.08735598,0.02078112,0.484218],"study_design_scores_gemma":[0.0001155924,0.0008270955,0.1760358,0.0005710403,0.0004484066,0.001631447,0.003499918,0.6446015,0.03589494,0.1126576,0.02295294,0.0007637604],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.455302,0.007789146,0.5112124,0.001401207,0.0005390292,0.0004069004,0.002096021,0.001698332,0.01955501],"genre_scores_gemma":[0.8990781,0.0009042195,0.09663117,0.0001318967,0.0001195755,0.0001567922,0.001561951,0.0002762302,0.001140012],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008020844,"threshold_uncertainty_score":0.0406214,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0262903710357365,"score_gpt":0.3489306135391119,"score_spread":0.3226402425033754,"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."}}