{"id":"W4392484188","doi":"10.1145/3636555.3636910","title":"Prompt-based and Fine-tuned GPT Models for Context-Dependent and -Independent Deductive Coding in Social Annotation","year":2024,"lang":"en","type":"article","venue":"","topic":"Innovative Teaching and Learning Methods","field":"Psychology","cited_by":31,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Computer science; Coding (social sciences); Annotation; Natural language processing; Artificial intelligence; Machine learning; Mathematics","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":[0.00186882,0.0001431134,0.0002115281,0.0002450723,0.00015752,0.00008115257,0.00004666144,0.0001210798,0.00006067498],"category_scores_gemma":[0.00009216683,0.0001291485,0.0000380824,0.0001700707,0.00006094736,0.0001144621,0.00001943508,0.0003361821,0.000003867418],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006637396,"about_ca_system_score_gemma":0.00004476656,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002066945,"about_ca_topic_score_gemma":0.00004925954,"domain_scores_codex":[0.9985645,0.0004266372,0.0002316364,0.0004288905,0.0001167838,0.0002314844],"domain_scores_gemma":[0.9993151,0.000465698,0.00004887322,0.00007253703,0.00006860345,0.00002921028],"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":[0.0006507584,0.0001982448,0.002624094,0.0002430519,0.0002431288,0.00003110161,0.05107573,0.0003624226,0.005402766,0.5655838,0.001754345,0.3718306],"study_design_scores_gemma":[0.02515657,0.002909491,0.1973795,0.0007756804,0.0004773517,0.0001427261,0.05299803,0.4691955,0.008470257,0.2258756,0.0134241,0.003195264],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5371292,0.0005356091,0.4553763,0.001178298,0.0004147826,0.0005886867,0.00001604678,0.000130956,0.004630169],"genre_scores_gemma":[0.9921999,0.000001049295,0.005183022,0.0001571788,0.0001307403,0.0001445369,0.00001471142,0.00002525483,0.002143595],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.468833,"threshold_uncertainty_score":0.5266522,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08746716021728913,"score_gpt":0.4109897229639487,"score_spread":0.3235225627466596,"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."}}