{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","sts","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.00166715,0.000170571,0.0004470039,0.0001697459,0.0002428727,0.00176266,0.003621121,0.00006042497,0.000001702067],"category_scores_gemma":[0.02128487,0.00009849531,0.00007786373,0.001094312,0.00453139,0.005684257,0.002301082,0.0001710501,3.66588e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002549114,"about_ca_system_score_gemma":0.0007336013,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003241178,"about_ca_topic_score_gemma":0.00001209186,"domain_scores_codex":[0.9983985,0.00002903648,0.0004677227,0.0003189335,0.0004793462,0.0003064703],"domain_scores_gemma":[0.9808285,0.0004568324,0.00064924,0.0003316397,0.0176622,0.00007158642],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"observational","study_design_scores_codex":[0.00003588895,0.00002634182,0.2933959,0.0003012588,0.00007476205,2.01375e-7,0.02745386,0.000002537599,0.0168261,0.6549702,0.000487477,0.006425414],"study_design_scores_gemma":[0.001713686,0.000337484,0.5904261,0.001340078,0.000176846,0.00001665644,0.01080207,0.003561125,0.2246777,0.158136,0.00838551,0.0004266905],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9830821,0.0003286929,0.00015997,0.007892306,0.0001367435,0.0002802439,0.000008475388,0.00002974867,0.008081773],"genre_scores_gemma":[0.9964957,0.00009474895,0.002979491,0.0001758692,0.00001992771,0.000009626656,2.127572e-7,0.00000397307,0.0002205107],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4968343,"threshold_uncertainty_score":0.9992736,"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."}}