{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01593187,0.0007352133,0.0005413613,0.003966011,0.002069227,0.002753021,0.001629945,0.002117896,0.002065143],"category_scores_gemma":[0.03639714,0.0003658248,0.001240346,0.002818521,0.001311146,0.002198932,0.001958652,0.001367942,0.0006561849],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001892825,"about_ca_system_score_gemma":0.001826594,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004994553,"about_ca_topic_score_gemma":0.009572297,"domain_scores_codex":[0.9877126,0.008591367,0.0005492602,0.001505673,0.001219483,0.000421608],"domain_scores_gemma":[0.9408857,0.04792039,0.002791366,0.003194148,0.003654336,0.001554062],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001463612,0.002620695,0.5159996,0.00197867,0.0004318944,0.01242213,0.1419687,0.0179236,0.01769572,0.009490085,0.01393665,0.2640687],"study_design_scores_gemma":[0.0003197746,0.001681483,0.2238715,0.0008552806,0.0004316753,0.01078191,0.1466767,0.4873743,0.02267651,0.02876107,0.07615642,0.0004134656],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8348505,0.0005637759,0.1492848,0.004210582,0.0001624202,0.001968241,0.003239149,0.0006709116,0.005049597],"genre_scores_gemma":[0.8474352,0.0002060184,0.1471473,0.0003210597,0.0001220264,0.0008573785,0.002300131,0.0001299925,0.001481017],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01593187,"threshold_uncertainty_score":0.08425671,"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."}}