{"id":"W4411377134","doi":"10.1038/s41597-025-05301-4","title":"The Austronesian and the Micronesian Comparative Dictionaries as CLDF datasets","year":2025,"lang":"en","type":"article","venue":"Scientific Data","topic":"Genomics and Phylogenetic Studies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University","funders":"Ministry of Education, India; Ministry of Education - Singapore","keywords":"Micronesian; Austronesian languages; Interoperability; Computer science; Comparative method; Linguistics; Comparative linguistics; Usability; Information retrieval; World Wide Web; History; Genealogy; Sociology; Philology","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":[0.004065047,0.000537205,0.0006114572,0.01339339,0.001674388,0.003199748,0.001979699,0.001111642,0.02175858],"category_scores_gemma":[0.02893027,0.0005164645,0.0006141766,0.01962877,0.001150242,0.003323723,0.006156095,0.001996515,0.01082377],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002804147,"about_ca_system_score_gemma":0.007159368,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02369901,"about_ca_topic_score_gemma":0.04935123,"domain_scores_codex":[0.9969928,0.0005455585,0.0009343072,0.0005706977,0.0007415506,0.0002150614],"domain_scores_gemma":[0.9842774,0.004447724,0.001715056,0.005561225,0.003031372,0.0009672327],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0009877781,0.0002334533,0.03306412,0.005870494,0.000177222,0.001682608,0.00928383,0.002840149,0.01469454,0.09117221,0.6199071,0.2200864],"study_design_scores_gemma":[0.00005272824,0.00001467639,0.01440241,0.0005953204,0.0000277469,0.000358551,0.001265661,0.0009782248,0.001577823,0.007905713,0.9727683,0.00005298097],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.03656388,0.002030991,0.02361214,0.002355618,0.00048634,0.0005419749,0.8983931,0.005322211,0.03069377],"genre_scores_gemma":[0.04401789,0.0008603535,0.08502169,0.0005240251,0.00005543014,0.00145384,0.8598603,0.001555455,0.006650951],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.02369901,"threshold_uncertainty_score":0.07278973,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0220802469138846,"score_gpt":0.3009350242765415,"score_spread":0.2788547773626569,"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."}}