{"id":"W4414693363","doi":"10.1109/tts.2025.3611984","title":"Using Large Language Models in Cluster Analysis in the Social Sciences","year":2025,"lang":"en","type":"article","venue":"IEEE Transactions on Technology and Society","topic":"Computational and Text Analysis Methods","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Institute on Governance","funders":"","keywords":"Cluster analysis; Metadata; Workflow; Context (archaeology); Variety (cybernetics); Exploratory data analysis; Thematic map; Topic model; Language model","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001521135,0.00005866156,0.0001417255,0.000555884,0.001060966,0.00003258072,0.0001734833,0.0001707096,0.00001443663],"category_scores_gemma":[0.000007010552,0.00004709126,0.0001395996,0.005513648,0.0004735687,0.00009378976,0.000002346276,0.000254168,3.807452e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005263942,"about_ca_system_score_gemma":0.00007251363,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006140465,"about_ca_topic_score_gemma":0.005330521,"domain_scores_codex":[0.9991436,0.0002329687,0.0001328009,0.0001766824,0.0001351712,0.0001787601],"domain_scores_gemma":[0.9996743,0.0002123697,0.0000264549,0.00006000367,0.00001980943,0.000007012733],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00002866185,0.0008971015,0.01127006,0.00002579066,0.000836516,0.000007308599,0.2478765,0.140154,0.0001985791,0.5270531,0.0002097677,0.07144263],"study_design_scores_gemma":[0.000934811,0.00002649982,0.004902654,0.0000262003,0.0005256957,7.892989e-7,0.2432977,0.5834265,0.0001845339,0.165891,0.0005090188,0.0002746204],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4119966,0.00009083365,0.5809925,0.005171625,0.00003579094,0.00007754474,0.000003381296,0.00002863437,0.001603125],"genre_scores_gemma":[0.9962158,0.00004062912,0.002789567,0.0007540939,0.000008530045,0.00001192187,3.388682e-7,0.000001259373,0.0001779288],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5842191,"threshold_uncertainty_score":0.8160198,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04261437470856897,"score_gpt":0.4023507107515543,"score_spread":0.3597363360429853,"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."}}