{"id":"W3217379076","doi":"10.1136/fmch-2021-001287","title":"Developing and testing an automated qualitative assistant (AQUA) to support qualitative analysis","year":2021,"lang":"en","type":"article","venue":"Family Medicine and Community Health","topic":"Qualitative Research Methods and Ethics","field":"Social Sciences","cited_by":47,"is_retracted":false,"has_abstract":true,"ca_institutions":"Centre for Family Medicine","funders":"Penn State College of Medicine; Huck Institutes of the Life Sciences; Pennsylvania State University; University of Pennsylvania","keywords":"Computer science; Latent Dirichlet allocation; Coding (social sciences); Machine learning; Information retrieval; Cluster analysis; Artificial intelligence; Transparency (behavior); Natural language processing; Data mining; Topic model; Statistics; Mathematics","routes":{"ca_aff":true,"ca_fund":false,"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":[0.08995779,0.001661094,0.001169632,0.003453569,0.002602855,0.004289208,0.003801582,0.001499099,0.02288491],"category_scores_gemma":[0.2520168,0.001620189,0.001508213,0.002278924,0.003043483,0.005367769,0.008285425,0.002948902,0.007987998],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.004131052,"about_ca_system_score_gemma":0.01351869,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003137762,"about_ca_topic_score_gemma":0.004505731,"domain_scores_codex":[0.9352953,0.04724921,0.004878067,0.006086636,0.005103778,0.001387129],"domain_scores_gemma":[0.631404,0.285304,0.009291939,0.02646392,0.04279624,0.004739874],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001356298,0.001638959,0.01705228,0.006192416,0.0002003136,0.001103297,0.1020309,0.007467842,0.02275744,0.02489296,0.03752304,0.7777843],"study_design_scores_gemma":[0.001754937,0.003374471,0.02297531,0.004694173,0.0003922957,0.002394058,0.09249263,0.2476918,0.09689238,0.09663329,0.4292013,0.001503379],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.06929451,0.0002641078,0.8735914,0.003033592,0.0006367484,0.02270845,0.004341216,0.01552266,0.01060738],"genre_scores_gemma":[0.05685857,0.000133385,0.9151531,0.0006153522,0.00004749716,0.0221723,0.001160605,0.0008531225,0.003006172],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.08995779,"threshold_uncertainty_score":0.4757479,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.7362791457125837,"score_gpt":0.6989436886705526,"score_spread":0.03733545704203112,"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."}}