{"id":"W4415323123","doi":"10.34105/j.kmel.2025.17.031","title":"From data to insights: Machine learning in thematic analysis of complex health conditions on social media","year":2025,"lang":"en","type":"article","venue":"Knowledge Management & E-Learning An International Journal","topic":"Data-Driven Disease Surveillance","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Thompson Rivers University","funders":"Thompson Rivers University","keywords":"Thematic analysis; Social media; Thematic map; Big data; Filter (signal processing); Process (computing); Class (philosophy); Focus (optics)","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.02905738,0.001289592,0.001096556,0.008208365,0.002171926,0.006753252,0.002412696,0.001653171,0.001967597],"category_scores_gemma":[0.08577071,0.0007395289,0.00252629,0.006262192,0.003053238,0.00693902,0.005043153,0.004310306,0.0007150358],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00323927,"about_ca_system_score_gemma":0.00263338,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004899048,"about_ca_topic_score_gemma":0.006607085,"domain_scores_codex":[0.9797323,0.01596287,0.0009223805,0.001779891,0.001269176,0.0003333388],"domain_scores_gemma":[0.8913638,0.09565955,0.004450165,0.003940281,0.00394247,0.0006436825],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005568479,0.0008215109,0.04163567,0.002244845,0.000664082,0.0006662594,0.02893021,0.07296155,0.0046158,0.04115717,0.0130966,0.7926495],"study_design_scores_gemma":[0.00005428672,0.0001087494,0.007859129,0.0005207274,0.0001001268,0.0001411338,0.0112477,0.7794494,0.002542209,0.1870541,0.01080028,0.0001221353],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07297285,0.001139061,0.9122705,0.006807912,0.0002230603,0.001238768,0.00136085,0.0009741283,0.003012856],"genre_scores_gemma":[0.2840424,0.0005262143,0.7115614,0.0005100056,0.0001842473,0.0012406,0.001207597,0.00010637,0.0006211366],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02905738,"threshold_uncertainty_score":0.1536719,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0916651111064284,"score_gpt":0.4145368601336094,"score_spread":0.322871749027181,"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."}}