{"id":"W4410705334","doi":"10.29173/cais2020","title":"Applying LLMs and Semantic Technologies for Data Extraction in Literature Reviews","year":2025,"lang":"fr","type":"article","venue":"Proceedings of the Annual Conference of CAIS / Actes du congrès annuel de l ACSI","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Computer science; Semantic technology; Data extraction; Information retrieval; Data science; Natural language processing; Semantic Web; Semantic computing; Political science; MEDLINE","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.05936283,0.001376266,0.001478957,0.01839021,0.001205895,0.006789868,0.001796019,0.001326138,0.004483581],"category_scores_gemma":[0.1977131,0.001165336,0.0042603,0.01395842,0.001025209,0.008322875,0.005403453,0.00124007,0.001649583],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.004448354,"about_ca_system_score_gemma":0.01241712,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006789122,"about_ca_topic_score_gemma":0.01351546,"domain_scores_codex":[0.95344,0.02796086,0.008804012,0.003175074,0.006044308,0.0005757759],"domain_scores_gemma":[0.7767392,0.1804354,0.01192563,0.01573327,0.0141719,0.0009946054],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002913289,0.0004668868,0.0368669,0.05212862,0.002553411,0.0006700823,0.02083955,0.01062793,0.01943775,0.02271518,0.02015813,0.8106224],"study_design_scores_gemma":[0.002348217,0.002528246,0.07973579,0.03119312,0.009259191,0.001808435,0.02612696,0.1620875,0.08074445,0.07870643,0.5244907,0.0009709913],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1452229,0.01200739,0.7154189,0.01157188,0.0006186148,0.02369237,0.04645439,0.02273744,0.0222762],"genre_scores_gemma":[0.1430417,0.002419336,0.8270658,0.001036531,0.00007152007,0.01149373,0.01296523,0.0004470315,0.001459181],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9406372,"threshold_uncertainty_score":0.3139444,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0450675097573206,"score_gpt":0.3213783706396935,"score_spread":0.2763108608823729,"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."}}