{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008721448,0.0000553024,0.00008679432,0.0001257402,0.0004455937,0.0001118609,0.0001022757,0.00004980716,0.00004400444],"category_scores_gemma":[0.0002188585,0.00005606179,0.00006315015,0.00036621,0.00001217252,0.0004550274,0.00001789482,0.00002977559,5.727249e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008878055,"about_ca_system_score_gemma":0.0001046963,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002127898,"about_ca_topic_score_gemma":0.002579195,"domain_scores_codex":[0.9992179,0.0001277572,0.0001908846,0.000158801,0.0001715578,0.0001331517],"domain_scores_gemma":[0.9994233,0.0002311434,0.00007813531,0.00009672376,0.0001465927,0.00002415284],"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":[0.00002212299,0.00006264834,0.002442474,0.00004330238,0.00002454237,8.796972e-8,0.002832951,0.6239496,0.0008201375,0.3537135,0.00008822425,0.01600042],"study_design_scores_gemma":[0.0001722505,0.00000345077,0.001678411,0.00001486281,0.0000240055,9.663577e-9,0.0006561677,0.9549295,0.0001002477,0.04184211,0.0005200722,0.00005892273],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0968025,0.00002169285,0.8988343,0.0006546561,0.00008416027,0.0002965896,0.000003582938,0.0000676011,0.003234929],"genre_scores_gemma":[0.8637785,6.124075e-7,0.1343838,0.000119636,0.00006762453,0.00001875499,0.00008759243,0.0000036887,0.001539812],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.766976,"threshold_uncertainty_score":0.3427192,"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."}}