{"id":"W4200456451","doi":"10.31234/osf.io/j76eg","title":"Challenges and Solutions for Data Analysis in an Adult Lifespan Study of over 100,000 Online Cognitive Test Completions","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Health, Environment, Cognitive Aging","field":"Environmental Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Baycrest Hospital","funders":"","keywords":"Computer science; Outlier; Visualization; Data mining; Smoothing; Data science; Data visualization; Information retrieval; Machine learning; Artificial intelligence","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.000966567,0.0003195367,0.0006834953,0.0002594928,0.0001841487,0.00004037985,0.0005136522,0.0001713166,0.0007493997],"category_scores_gemma":[0.0006336267,0.0003365165,0.00007547098,0.0003718289,0.0002771933,0.000316583,0.003705754,0.0004660797,0.000005091274],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001679151,"about_ca_system_score_gemma":0.0000440509,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.00980975,"about_ca_topic_score_gemma":0.3608017,"domain_scores_codex":[0.9962568,0.0003973183,0.0006516398,0.001819112,0.000410165,0.0004649775],"domain_scores_gemma":[0.9972364,0.0008844066,0.0002821992,0.001344188,0.0000467216,0.0002061133],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00003664394,0.01352312,0.9332079,0.0001648792,0.0007452866,0.00001691438,0.01034027,0.003775328,0.0001923073,0.00003167844,0.00003573342,0.03792996],"study_design_scores_gemma":[0.0008892471,0.0002659314,0.9132137,0.00009347746,0.0008534501,9.074852e-7,0.02552776,0.05873398,0.000005631676,0.00007216907,0.00003148958,0.000312198],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9875098,0.0002804314,0.005687249,0.0005342691,0.00003353115,0.001988975,0.003249695,0.00002871669,0.0006873091],"genre_scores_gemma":[0.9914163,0.001704405,0.002939769,0.0002209237,0.00003500128,0.0001899766,0.003425858,0.00002899099,0.00003871926],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.350992,"threshold_uncertainty_score":0.9999087,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1780504610705116,"score_gpt":0.377899496763216,"score_spread":0.1998490356927044,"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."}}