{"id":"W4399901273","doi":"10.2196/50918","title":"Sentiment analysis and natural language processing using Reddit data to evaluate patient opinions on hair loss therapeutics (Preprint)","year":2023,"lang":"en","type":"article","venue":"JMIR Dermatology","topic":"Social Media in Health Education","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"","keywords":"Preprint; Natural (archaeology); Computer science; Sentiment analysis; Natural language processing; Data science; Artificial intelligence; World Wide Web; History; Archaeology","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":[0.0004093355,0.00009522976,0.000203802,0.000292119,0.0005108233,0.00005744919,0.00026114,0.0001007864,0.00002735854],"category_scores_gemma":[0.0004201937,0.00009571343,0.00002980755,0.001518414,0.0001358943,0.0001210263,0.0001792203,0.0001476315,0.00005691333],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001952838,"about_ca_system_score_gemma":0.0003116178,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001107858,"about_ca_topic_score_gemma":0.001045321,"domain_scores_codex":[0.9982758,0.0004060162,0.0002479571,0.000368223,0.000336656,0.0003653794],"domain_scores_gemma":[0.9987718,0.0005186519,0.0001234456,0.0003684628,0.00008850128,0.0001291788],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"observational","study_design_scores_codex":[0.00006530841,0.0001896203,0.222922,0.0001741413,0.0003823761,0.00006177435,0.4999346,0.0004664601,0.0003969449,0.0005547063,0.002110271,0.2727419],"study_design_scores_gemma":[0.000952348,0.0001213575,0.6524541,0.0002742395,0.0007585856,0.00004087596,0.1141497,0.1407118,0.0009766266,0.0005604748,0.08794513,0.001054818],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9841403,0.0002179146,0.00004633697,0.01303415,0.00130335,0.001041777,0.00001916287,0.0001230681,0.00007396593],"genre_scores_gemma":[0.9981567,0.00003063285,0.0005361805,0.0007557017,0.0001625622,0.0002254282,0.0001063502,0.00001068468,0.00001578262],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4295321,"threshold_uncertainty_score":0.3928891,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1695240685657724,"score_gpt":0.4939700251387626,"score_spread":0.3244459565729902,"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."}}