{"id":"W4254367530","doi":"10.2196/preprints.13296","title":"Using Natural Language Processing to Examine the Uptake, Content, and Readability of Media Coverage of a Pan-Canadian Drug Safety Research Project: Cross-Sectional Observational Study (Preprint)","year":2019,"lang":"en","type":"preprint","venue":"","topic":"Health Literacy and Information Accessibility","field":"Health Professions","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Saskatchewan; Nova Scotia Health Authority; Dalhousie University","funders":"","keywords":"Readability; Observational study; Sentence; Computer science; Natural language processing; Medicine; Information retrieval; Pathology","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.003969007,0.0003714632,0.0005100982,0.007798197,0.001677247,0.002251258,0.001151735,0.0006931885,0.002680555],"category_scores_gemma":[0.02534706,0.0004224179,0.0009852584,0.01070414,0.0009806667,0.001252565,0.001268883,0.0006978106,0.0005130481],"about_ca_system_candidate":true,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.009747701,"about_ca_system_score_gemma":0.01183387,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.8343207,"about_ca_topic_score_gemma":0.7721708,"domain_scores_codex":[0.9962456,0.0006427948,0.0007636621,0.0005394761,0.001410053,0.0003984369],"domain_scores_gemma":[0.9732231,0.007582338,0.009637007,0.0008564223,0.007661235,0.001039961],"domain_codex":null,"domain_gemma":"reporting","domain_candidate":"reporting","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0002087154,0.0001479762,0.9746201,0.0006609757,0.0001864863,0.0001758669,0.009937417,0.00006960626,0.0002819753,0.0001045065,0.002864492,0.01074192],"study_design_scores_gemma":[0.00001203795,0.00005758286,0.9935318,0.00008715996,0.00005295198,0.00009996628,0.004115379,0.0002430055,0.0001070221,0.0000236396,0.001644343,0.00002499932],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9712229,0.0007533392,0.0002963923,0.0004235792,0.00001491872,0.0005381675,0.02397412,0.00003011205,0.002746332],"genre_scores_gemma":[0.9838253,0.0008056962,0.001544481,0.0003580069,0.00003537244,0.000836757,0.01115055,0.00002394017,0.001419839],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9977487,"threshold_uncertainty_score":0.3333101,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4866557656138405,"score_gpt":0.5579563756532481,"score_spread":0.07130061003940757,"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."}}