{"id":"W4410764418","doi":"10.2196/67333","title":"Differential Analysis of Age, Gender, Race, Sentiment, and Emotion in Substance Use Discourse on Twitter During the COVID-19 Pandemic: A Natural Language Processing Approach","year":2025,"lang":"en","type":"article","venue":"JMIR Infodemiology","topic":"Mental Health via Writing","field":"Psychology","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Substance Abuse and Mental Health Services Administration","keywords":"Social media; Pandemic; Ethnic group; Demographics; Psychology; Focus group; Coronavirus disease 2019 (COVID-19); Demography; Computer science; Medicine; Political science; World Wide Web; Sociology; Disease","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"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":[0.0005097706,0.000189732,0.0005002351,0.0004886585,0.0001336003,0.00002012751,0.0001784559,0.0001933575,0.00004926056],"category_scores_gemma":[0.00009396291,0.0001418812,0.00008768071,0.0006583494,0.0002370496,0.00008439186,0.00009639245,0.0005238199,0.000001481056],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001830971,"about_ca_system_score_gemma":0.00003116383,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003631056,"about_ca_topic_score_gemma":0.0003869015,"domain_scores_codex":[0.9977458,0.0006688501,0.0006018634,0.0004620421,0.0001115097,0.000409975],"domain_scores_gemma":[0.9988068,0.0004722867,0.000310123,0.000320838,0.00001966364,0.00007027013],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0004142581,0.0001579192,0.9757662,0.0002899577,0.0001859988,0.00001589214,0.02005504,0.0003266014,0.001000293,0.0006896883,0.00007472782,0.001023367],"study_design_scores_gemma":[0.001298022,0.00002029823,0.9853244,0.00004005189,0.0001223656,0.00001564475,0.005931738,0.007018316,0.00001648705,0.00006918899,0.0000186545,0.0001248037],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9963795,0.0005521095,0.001449965,0.0002373826,0.000141433,0.000608754,0.0000139235,0.0000437225,0.0005732392],"genre_scores_gemma":[0.9973494,0.00001715996,0.0001489998,0.001535233,0.00003422934,0.0001769886,0.00006049255,0.00001055668,0.0006669853],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.0141233,"threshold_uncertainty_score":0.5785747,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09432672901261528,"score_gpt":0.4434785616981538,"score_spread":0.3491518326855385,"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."}}