{"id":"W4401506443","doi":"10.1073/pnas.2308950121","title":"GPT is an effective tool for multilingual psychological text analysis","year":2024,"lang":"en","type":"article","venue":"Proceedings of the National Academy of Sciences","topic":"Misinformation and Its Impacts","field":"Social Sciences","cited_by":235,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institute of Mental Health; Natural Sciences and Engineering Research Council of Canada; Templeton World Charity Foundation; Sage Foundation; Russell Sage Foundation","keywords":"Computer science; Natural language processing; Artificial intelligence; Coding (social sciences); Text mining; Sentiment analysis; Information retrieval","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003231578,0.00006542849,0.0001314615,0.0002779287,0.0003928008,0.0001200572,0.0005643548,0.00009695733,0.00009817166],"category_scores_gemma":[0.001435932,0.00004057487,0.000156536,0.001885847,0.0008049182,0.0009700917,0.00003325979,0.0001016812,0.000002770686],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004017234,"about_ca_system_score_gemma":0.00003809413,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001541882,"about_ca_topic_score_gemma":4.895606e-7,"domain_scores_codex":[0.9983313,0.00001212507,0.0002502842,0.000198876,0.001048842,0.0001585539],"domain_scores_gemma":[0.9992366,0.000301856,0.0001650547,0.000006157015,0.000241131,0.00004923423],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"observational","study_design_scores_codex":[0.00007905676,0.000221466,0.00428471,0.0001660307,0.000289719,8.485588e-9,0.092614,0.0002460114,0.02537862,0.7907705,0.01053452,0.07541541],"study_design_scores_gemma":[0.0009053298,0.0007219671,0.4277509,0.0002895255,0.0005250988,0.000005373216,0.02914231,0.0617939,0.149185,0.3030183,0.02593787,0.0007244216],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9762404,0.00005758819,0.00001582275,0.003251781,0.00004065683,0.0003557405,0.00003441185,0.00002973722,0.01997389],"genre_scores_gemma":[0.9977345,0.00001661381,0.001080945,0.0005910154,0.0001228541,0.000007589817,1.903205e-7,0.000001761318,0.0004445117],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4877522,"threshold_uncertainty_score":0.3021146,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08362648693599621,"score_gpt":0.4454008735722634,"score_spread":0.3617743866362672,"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."}}