{"id":"W4319597719","doi":"10.5334/johd.95","title":"MultiHATHI: A Complete Collection of Multilingual Prose Fiction in the HathiTrust Digital Library","year":2023,"lang":"en","type":"article","venue":"Journal of Open Humanities Data","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Bespoke; Metadata; Computer science; Digital library; Classifier (UML); English language; World Wide Web; Information retrieval; Natural language processing; Artificial intelligence; Linguistics; Literature; Art; Poetry","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":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0009355603,0.00009479409,0.0001961448,0.0003122716,0.0001407586,0.00155337,0.005260484,0.00004035061,0.00001584746],"category_scores_gemma":[0.0002838131,0.00006512127,0.00003520894,0.0006198023,0.00007215431,0.007569914,0.002062877,0.0003140186,0.000004577105],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002493111,"about_ca_system_score_gemma":0.0001463658,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004781633,"about_ca_topic_score_gemma":0.00002171994,"domain_scores_codex":[0.9987667,0.0001044829,0.0004644383,0.0001698661,0.0003583296,0.0001361551],"domain_scores_gemma":[0.9986027,0.000227359,0.0004401445,0.0006062138,0.0001047726,0.00001875006],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0008370737,0.001578193,0.006042131,0.0006639972,0.0002472283,0.001501937,0.09623351,0.0001176216,0.003862146,0.0338819,0.7028102,0.1522241],"study_design_scores_gemma":[0.01256857,0.005299632,0.03168611,0.006282806,0.0002091466,0.005073335,0.04883227,0.2584504,0.02969742,0.1609772,0.4378755,0.003047566],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8269555,0.01240023,0.09231337,0.01548546,0.005837437,0.008461528,0.004531899,0.00219767,0.03181691],"genre_scores_gemma":[0.92865,0.000098731,0.06989359,0.0002677996,0.0001913635,0.000008718805,0.0001670281,0.000018901,0.0007038974],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2649346,"threshold_uncertainty_score":0.9994831,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1386274117001172,"score_gpt":0.3461929316426599,"score_spread":0.2075655199425427,"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."}}