{"id":"W4226294230","doi":"10.2196/35446","title":"Unsupervised Machine Learning to Detect and Characterize Barriers to Pre-exposure Prophylaxis Therapy: Multiplatform Social Media Study","year":2022,"lang":"en","type":"article","venue":"JMIR Infodemiology","topic":"HIV/AIDS Research and Interventions","field":"Medicine","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"U.S. Department of Health and Human Services","keywords":"Ethnic group; Social media; Thematic analysis; Medicine; Human immunodeficiency virus (HIV); Sexual orientation; Psychology; Medical education; Artificial intelligence; Family medicine; Computer science; Qualitative research; World Wide Web; Social psychology; Political science","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0009458206,0.0002020584,0.0004999568,0.0003367832,0.0005557484,0.00001999203,0.0002170532,0.0001019287,0.001599781],"category_scores_gemma":[0.001850607,0.0001708504,0.0001318639,0.0003721702,0.00007372985,0.00007013245,0.0004832452,0.000978957,0.00002813069],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001209538,"about_ca_system_score_gemma":0.0001236358,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001583754,"about_ca_topic_score_gemma":0.0001701184,"domain_scores_codex":[0.9978633,0.0004789225,0.0004373386,0.0004083159,0.0003096085,0.0005025411],"domain_scores_gemma":[0.9987608,0.0002375105,0.00007150334,0.0002309147,0.00009625498,0.0006030794],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.006304386,0.0003661579,0.6740052,0.0000720763,0.0002960423,0.00007926617,0.08403176,0.00005052886,0.03832261,0.0000274229,0.002630594,0.193814],"study_design_scores_gemma":[0.004976161,0.04541994,0.9141195,0.00001879179,0.00002275556,0.0000463396,0.007762379,0.0009364311,0.0001618763,0.00006003065,0.02617124,0.0003045227],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9955525,0.0001164264,0.00007300911,0.001438931,0.00007260519,0.002375396,0.00008500474,0.0001040548,0.0001820121],"genre_scores_gemma":[0.9955623,0.00001708746,0.0002518642,0.0009802575,0.0001464608,0.001996167,0.00007535615,0.00003111011,0.0009394489],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2401144,"threshold_uncertainty_score":0.9993129,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03485767340195346,"score_gpt":0.3390584428319096,"score_spread":0.3042007694299562,"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."}}