{"id":"W2808903237","doi":"10.2196/medinform.9162","title":"Novel Approach to Cluster Patient-Generated Data Into Actionable Topics: Case Study of a Web-Based Breast Cancer Forum","year":2018,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Social Media in Health Education","field":"Social Sciences","cited_by":45,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Social media; Topic model; Categorization; Computer science; Context (archaeology); Cluster analysis; Breast cancer; World Wide Web; Data science; Information retrieval; Medicine; Artificial intelligence; Cancer","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":[0.0007110041,0.0001155226,0.0002170286,0.0001318809,0.0004665274,0.00004527369,0.0005424483,0.0002067236,0.0002127628],"category_scores_gemma":[0.001143196,0.0001030403,0.00002057129,0.000768171,0.0002751135,0.0004409289,0.0001935147,0.0002154238,0.00002054122],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002846744,"about_ca_system_score_gemma":0.002191518,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.02691304,"about_ca_topic_score_gemma":0.02295356,"domain_scores_codex":[0.9975074,0.0001264949,0.0006899321,0.0001536819,0.001162778,0.0003597424],"domain_scores_gemma":[0.9979737,0.0003001923,0.0002597556,0.0004635447,0.0004998535,0.0005029487],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"qualitative","study_design_scores_codex":[0.00007404216,0.002615019,0.009331477,0.000294485,0.00005350956,0.000004181698,0.7778779,0.00004211552,0.000007513532,0.0002217433,0.03658438,0.1728936],"study_design_scores_gemma":[0.002224345,0.0004536565,0.001121676,0.0001301545,0.00004430298,0.00003192335,0.7893399,0.06840318,0.00002168312,0.0000198445,0.1378834,0.0003259032],"study_design_candidate":"qualitative","study_design_consensus":"qualitative","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.987246,0.000009758041,0.001317044,0.004696518,0.002643384,0.00266526,0.00005144331,0.00006320345,0.001307345],"genre_scores_gemma":[0.9882849,0.000005998494,0.005848079,0.003127556,0.001431617,0.001217187,0.00004904428,0.00001234856,0.00002329816],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1725677,"threshold_uncertainty_score":0.994875,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1048345991082292,"score_gpt":0.4222176527661611,"score_spread":0.317383053657932,"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."}}