{"id":"W4412022751","doi":"10.37236/12851","title":"Characterizations of Perfectly Clustering Words","year":2025,"lang":"en","type":"article","venue":"The Electronic Journal of Combinatorics","topic":"Advanced Algebra and Logic","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université du Québec à Montréal; Université de Moncton","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Cluster analysis; Natural language processing; Computer science; Artificial intelligence; Mathematics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0005328976,0.0006710009,0.0005374293,0.001562948,0.001914284,0.001815148,0.001087515,0.001256347,0.007745071],"category_scores_gemma":[0.002992345,0.0004383917,0.0006895646,0.001522956,0.002017719,0.004522982,0.001264378,0.001284335,0.001336932],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001027027,"about_ca_system_score_gemma":0.0005513232,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0007327261,"about_ca_topic_score_gemma":0.0006371967,"domain_scores_codex":[0.9983702,0.0002213933,0.0001334647,0.0005070161,0.0004471121,0.000320883],"domain_scores_gemma":[0.9975759,0.0006338293,0.0005043049,0.0004716578,0.0005731371,0.0002410617],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0001087777,0.00003718696,0.0008630881,0.0001105788,0.00001367288,0.0004219857,0.0005195961,0.004259305,0.01229496,0.9563169,0.001689452,0.02336452],"study_design_scores_gemma":[0.00001895463,0.00014157,0.000822042,0.00004796213,0.00002239667,0.001430981,0.0005168335,0.02052893,0.01058017,0.9405931,0.02521033,0.00008671793],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4252819,0.001544628,0.5056345,0.0006246989,0.0002999122,0.0001702355,0.0007721061,0.0009132765,0.06475869],"genre_scores_gemma":[0.8953801,0.0004587107,0.08737987,0.0003785536,0.0002942172,0.0001555935,0.001013467,0.0002353329,0.01470422],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007745071,"threshold_uncertainty_score":0.02590978,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.004797784448982438,"score_gpt":0.2333496507858884,"score_spread":0.2285518663369059,"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."}}