{"id":"W4412568039","doi":"10.1109/ichi64645.2025.00032","title":"Privacy Preservation Embedding-Based Clustering for Population Stratification Using Large Language Models","year":2025,"lang":"en","type":"article","venue":"","topic":"Computational and Text Analysis Methods","field":"Social Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Windsor","funders":"Natural Sciences and Engineering Research Council of Canada; Mitacs","keywords":"Cluster analysis; Computer science; Embedding; Stratification (seeds); Population stratification; Data mining; Artificial intelligence","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":[0.004745387,0.0008050757,0.0009583164,0.0008760589,0.001018641,0.001891811,0.001313846,0.001242514,0.001341331],"category_scores_gemma":[0.02177014,0.0004641843,0.00120666,0.001145035,0.00194152,0.004556366,0.003883309,0.002477487,0.0006243411],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00118277,"about_ca_system_score_gemma":0.001677583,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001797464,"about_ca_topic_score_gemma":0.001965833,"domain_scores_codex":[0.9950709,0.002556375,0.0002966243,0.0008518372,0.0009195206,0.0003046903],"domain_scores_gemma":[0.9859511,0.007604009,0.001012207,0.004395127,0.0007975734,0.0002399147],"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":[0.0005971425,0.0002256189,0.007271907,0.0002338406,0.0002181807,0.0005609634,0.001150855,0.534754,0.01087685,0.2220784,0.005845914,0.2161864],"study_design_scores_gemma":[0.00001737933,0.00004913389,0.0003206023,0.00001101202,0.0000164522,0.0001457387,0.00007175509,0.8922253,0.0035098,0.1024612,0.001150789,0.00002085494],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03475154,0.0002115365,0.9625036,0.0006864595,0.00003496877,0.0000610585,0.0002549761,0.0005986767,0.0008972284],"genre_scores_gemma":[0.7586,0.000364733,0.2366257,0.0004314627,0.00008969385,0.0002035395,0.0009432851,0.0001619776,0.002579687],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004745387,"threshold_uncertainty_score":0.02509624,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09104518065992609,"score_gpt":0.4524024797789539,"score_spread":0.3613572991190278,"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."}}