{"id":"W3214120707","doi":"10.1075/cilt.356.09emb","title":"The perils of big data","year":2021,"lang":"en","type":"book-chapter","venue":"Amsterdam studies in the theory and history of linguistic science. Series 4, Current issues in linguistic theory","topic":"Artificial Intelligence in Healthcare","field":"Health Professions","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"York University","funders":"","keywords":"Big data; Data science; Process (computing); Computer science; History; Data mining","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":[0.06599303,0.0007564641,0.001300388,0.004879614,0.004359715,0.02236521,0.004009672,0.007692735,0.01020442],"category_scores_gemma":[0.1163038,0.0007247287,0.001083934,0.008765613,0.01941893,0.04650563,0.009382768,0.0206008,0.003480115],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.004220855,"about_ca_system_score_gemma":0.006974851,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002212455,"about_ca_topic_score_gemma":0.00241797,"domain_scores_codex":[0.9518819,0.02585627,0.001380621,0.002117112,0.01761591,0.001148102],"domain_scores_gemma":[0.7397544,0.2207584,0.00323173,0.01854089,0.01290647,0.004808165],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00006658587,0.00002929654,0.0007483321,0.0006833906,0.00005601256,0.000152493,0.002030805,0.0006335622,0.0001305736,0.7059878,0.2234854,0.06599571],"study_design_scores_gemma":[0.0000137371,0.00002132291,0.000469104,0.001459196,0.00001266121,0.0001942218,0.001962965,0.001118508,0.0001490813,0.5026327,0.4919311,0.00003540681],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"commentary","genre_gemma":"other","genre_scores_codex":[0.0024921,0.09199673,0.01818523,0.8269334,0.01532757,0.00006060136,0.0005602628,0.0003346408,0.04410941],"genre_scores_gemma":[0.2073525,0.2013423,0.04991385,0.421206,0.07751328,0.0008828514,0.001735383,0.001885167,0.03816871],"genre_candidate":"other","genre_consensus":null,"teacher_disagreement_score":0.06599303,"threshold_uncertainty_score":0.3490087,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3078001384030181,"score_gpt":0.4859037882114023,"score_spread":0.1781036498083842,"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."}}