{"id":"W4389565699","doi":"10.3233/faia230965","title":"Using Large Language Models to Support Thematic Analysis in Empirical Legal Studies","year":2023,"lang":"en","type":"book-chapter","venue":"Frontiers in artificial intelligence and applications","topic":"Law in Society and Culture","field":"Social Sciences","cited_by":16,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal; Research Unit on Children's Psychosocial Maladjustment","funders":"","keywords":"Thematic analysis; Coding (social sciences); Thematic map; Computer science; Qualitative analysis; Empirical research; Qualitative research; Quality (philosophy); Data science; Sociology; Epistemology; Social science; Geography; Cartography","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":[0.0006816909,0.0001835203,0.0004755539,0.0001634204,0.0003658168,0.00008103115,0.0002693822,0.0002867024,0.00005956321],"category_scores_gemma":[0.00003882728,0.0001794348,0.0001579347,0.0008436491,0.0003086156,0.0001246242,0.00009398923,0.0003043669,0.0000481462],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001882881,"about_ca_system_score_gemma":0.0001054691,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002523162,"about_ca_topic_score_gemma":0.01207699,"domain_scores_codex":[0.998431,0.00004063568,0.0004720292,0.0004325416,0.0002970644,0.0003267544],"domain_scores_gemma":[0.9994013,0.00008463253,0.0001047985,0.0002308468,0.00007809848,0.0001003683],"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":[0.000005794722,0.00005651439,0.0001511793,0.00003116754,0.0002242561,0.00001012851,0.0822003,0.002750524,0.000002634176,0.89923,0.003137491,0.01219998],"study_design_scores_gemma":[0.00001953102,0.0000156703,0.000004753012,0.0001010811,0.0003077668,3.08527e-7,0.188894,0.005432501,0.000015779,0.7606411,0.04416651,0.0004009782],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"other","genre_scores_codex":[0.002333691,0.006018298,0.6418775,0.004905695,0.0009905759,0.004612683,0.0004864974,0.0003506216,0.3384244],"genre_scores_gemma":[0.2702499,0.02151454,0.03512458,0.00300949,0.002565881,0.00172119,0.0003717404,0.00023022,0.6652125],"genre_candidate":"other","genre_consensus":null,"teacher_disagreement_score":0.6067529,"threshold_uncertainty_score":0.731714,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2110122329014868,"score_gpt":0.4261866455256179,"score_spread":0.2151744126241311,"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."}}