{"id":"W4416185268","doi":"10.1016/j.vhri.2025.101539","title":"Evaluating the Performance of Claude 3.7 Sonnet in Data Extraction Automation for Systematic Literature Reviews","year":2025,"lang":"en","type":"article","venue":"Value in Health Regional Issues","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":1,"is_retracted":false,"has_abstract":false,"ca_institutions":"EVRAZ (Canada)","funders":"","keywords":"Systematic review; Sonnet; Data extraction; Automation; Extraction (chemistry)","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":[],"consensus_categories":[],"category_scores_codex":[0.002990853,0.00007854283,0.0002407209,0.00006768915,0.00005162461,0.00001067803,0.0002532798,0.0001098557,6.052317e-7],"category_scores_gemma":[0.001066193,0.0000516301,0.00002986581,0.0001996817,0.00004636489,0.000006748832,0.0000528079,0.00009813479,5.516483e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002581292,"about_ca_system_score_gemma":0.0001250631,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008381556,"about_ca_topic_score_gemma":0.00006245842,"domain_scores_codex":[0.9986071,0.0003824949,0.0005460347,0.0002154387,0.0001128579,0.000136072],"domain_scores_gemma":[0.9991394,0.0001874384,0.0002366338,0.0003750641,0.00004621613,0.00001524478],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"systematic_review","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001547222,0.0009081853,0.017268,0.6108872,0.0002039078,0.00000319061,0.008853368,0.0125231,0.03967467,0.007002148,0.07929127,0.2218378],"study_design_scores_gemma":[0.001931252,0.001603343,0.02019091,0.1422073,0.00004534081,0.00002853995,0.001247983,0.713173,0.002560687,0.001393771,0.1152263,0.0003915702],"study_design_candidate":"systematic_review","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7361778,0.2500685,0.0009606372,0.01044051,0.000270023,0.002027127,0.0000209015,0.00001400048,0.00002053703],"genre_scores_gemma":[0.9304437,0.03496192,0.03043903,0.001595932,0.0003348134,0.0005777909,0.0007702571,0.00001644407,0.0008601266],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7006499,"threshold_uncertainty_score":0.2105414,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1678273020486737,"score_gpt":0.4849505993754512,"score_spread":0.3171232973267775,"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."}}