{"id":"W4410705321","doi":"10.29173/cais1943","title":"Investigating Document Type Discrepancies between OpenAlex and the Web of Science","year":2025,"lang":"en","type":"article","venue":"Proceedings of the Annual Conference of CAIS / Actes du congrès annuel de l ACSI","topic":"Semantic Web and Ontologies","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa; Dalhousie University","funders":"","keywords":"World Wide Web; Web of science; Type (biology); Computer science; Information retrieval; Data science; MEDLINE; Political science; Biology","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","bibliometrics"],"consensus_categories":[],"category_scores_codex":[0.05204695,0.0004822263,0.001011245,0.02721222,0.00164247,0.01182503,0.001785008,0.001480722,0.003225806],"category_scores_gemma":[0.3131177,0.000492025,0.0008334856,0.03913007,0.003144545,0.009655122,0.008314976,0.00113868,0.001823226],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002781064,"about_ca_system_score_gemma":0.002330643,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006232139,"about_ca_topic_score_gemma":0.004858789,"domain_scores_codex":[0.9187101,0.01843937,0.01430856,0.007462522,0.03908827,0.001991277],"domain_scores_gemma":[0.485268,0.2715342,0.1053173,0.04845897,0.08679021,0.002631428],"domain_codex":null,"domain_gemma":"evaluation","domain_candidate":"evaluation","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.000631634,0.0001141932,0.7964613,0.001940259,0.0004950198,0.0005213036,0.02659387,0.001629983,0.002312329,0.01861266,0.005731893,0.1449556],"study_design_scores_gemma":[0.00004504675,0.0001825096,0.8204519,0.00225558,0.0003760236,0.00181312,0.03195373,0.008103467,0.01067967,0.02664866,0.09729919,0.0001911202],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9259405,0.006247791,0.02142025,0.002549728,0.0003624748,0.000259332,0.0124105,0.0005776358,0.03023175],"genre_scores_gemma":[0.9717058,0.001297484,0.01303346,0.0003125485,0.0001696751,0.000195537,0.01067287,0.0002912479,0.002321302],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9727878,"threshold_uncertainty_score":0.2752539,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02350689428926182,"score_gpt":0.2704189824575527,"score_spread":0.2469120881682909,"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."}}