{"id":"W2795677445","doi":"10.1080/01639374.2018.1438551","title":"Dealing with False Friends to Avoid Errors in Subject Analysis in Slavic Cataloging: An Overview of Resources and Strategies","year":2018,"lang":"en","type":"article","venue":"Cataloging & Classification Quarterly","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Cataloging; Slavic languages; Subject (documents); Workflow; Computer science; World Wide Web; Linguistics; Meaning (existential); Psychology; Database","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":[0.05183155,0.00233495,0.002573737,0.02928958,0.009293324,0.01745351,0.006835845,0.003551537,0.009775029],"category_scores_gemma":[0.1780091,0.002114467,0.001593659,0.01497444,0.008726414,0.03055256,0.01753163,0.003835927,0.01438924],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003889624,"about_ca_system_score_gemma":0.01137441,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008232329,"about_ca_topic_score_gemma":0.009923766,"domain_scores_codex":[0.937604,0.02881102,0.007549697,0.005635468,0.01831357,0.002086228],"domain_scores_gemma":[0.8086567,0.08242824,0.02109789,0.04224548,0.04216114,0.003410609],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0002219961,0.0001869701,0.0136255,0.00188399,0.000118762,0.00146458,0.04566714,0.0006927682,0.00365292,0.03311453,0.05169894,0.8476719],"study_design_scores_gemma":[0.00007034811,0.0001848492,0.01358962,0.008657952,0.0003734742,0.009348541,0.07100599,0.008666369,0.02183906,0.1275125,0.7380989,0.0006525067],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"review","genre_scores_codex":[0.0565131,0.02482393,0.8041363,0.02624423,0.001967083,0.001804239,0.0009011775,0.01917002,0.06443986],"genre_scores_gemma":[0.204428,0.01328923,0.7350091,0.004485643,0.001175923,0.001145181,0.002582214,0.00529935,0.03258533],"genre_candidate":"review","genre_consensus":null,"teacher_disagreement_score":0.05183155,"threshold_uncertainty_score":0.2741147,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04421775578432104,"score_gpt":0.3324192626887967,"score_spread":0.2882015069044757,"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."}}