{"id":"W4290994493","doi":"10.2196/38936","title":"Standardized Description of the Feature Extraction Process to Transform Raw Data Into Meaningful Information for Enhancing Data Reuse: Consensus Study","year":2022,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Electronic Health Records Systems","field":"Health Professions","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Raw data; Data mining; Data warehouse; Schema (genetic algorithms); Information retrieval; Reuse; Data science; Database; Engineering","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.170004,0.002285664,0.002544382,0.01718642,0.002735753,0.009470954,0.007751003,0.003900901,0.004111282],"category_scores_gemma":[0.2182456,0.001538071,0.005817384,0.01466764,0.004117787,0.01100811,0.006766255,0.005414429,0.002443609],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.008420873,"about_ca_system_score_gemma":0.03539149,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007714569,"about_ca_topic_score_gemma":0.00732386,"domain_scores_codex":[0.8607259,0.0666984,0.04327981,0.006251588,0.02145849,0.001585759],"domain_scores_gemma":[0.7735214,0.08946413,0.01605341,0.03328386,0.08547688,0.0022004],"domain_codex":null,"domain_gemma":"reproducibility","domain_candidate":"reproducibility","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"qualitative","study_design_scores_codex":[0.0002865947,0.000629047,0.0102216,0.02997307,0.001284174,0.0009672414,0.01300606,0.008769015,0.006279301,0.137627,0.0417828,0.7491741],"study_design_scores_gemma":[0.000350488,0.001077057,0.01652662,0.0920451,0.002274222,0.003122888,0.0178845,0.02793376,0.02327389,0.1055311,0.7092525,0.0007278601],"study_design_candidate":"qualitative","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.009805566,0.01596348,0.9404837,0.00960639,0.001097215,0.008478301,0.002469624,0.001027706,0.01106791],"genre_scores_gemma":[0.04045542,0.01036469,0.9308172,0.001681516,0.000215409,0.008927612,0.005666126,0.000300144,0.001571777],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.829996,"threshold_uncertainty_score":0.8990777,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1026964901352669,"score_gpt":0.4815325228663681,"score_spread":0.3788360327311012,"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."}}