{"id":"W3041134523","doi":"10.1080/17517575.2020.1790043","title":"Valuing free-form text data from maintenance logs through transfer learning with CamemBERT","year":2020,"lang":"en","type":"article","venue":"Enterprise Information Systems","topic":"Imbalanced Data Classification Techniques","field":"Computer Science","cited_by":36,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Computer science; Exploit; Transfer of learning; Artificial intelligence; Machine learning; Scheduling (production processes); Engineering","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":[],"consensus_categories":[],"category_scores_codex":[0.0003416463,0.000221831,0.000269354,0.00007050446,0.0001540853,0.0006626573,0.00286414,0.00008799947,0.00002132877],"category_scores_gemma":[0.0001515151,0.0001828406,0.00003559177,0.000399558,0.00004675172,0.01078586,0.0005399541,0.0003180252,0.0004504833],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008956652,"about_ca_system_score_gemma":0.00008525085,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003711245,"about_ca_topic_score_gemma":0.000007044412,"domain_scores_codex":[0.997909,0.00008874078,0.0007024545,0.000397968,0.0006041924,0.0002976483],"domain_scores_gemma":[0.9976687,0.00009258762,0.0002664319,0.001683479,0.0001729986,0.0001157329],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0007637735,0.0002813554,0.02390289,0.001724442,0.000809936,0.0001259564,0.2655079,0.01414224,0.005175533,0.135889,0.4479507,0.1037262],"study_design_scores_gemma":[0.001087654,0.0001343132,0.0006510874,0.0004315218,0.00001434817,0.00004097716,0.001745638,0.5734867,0.001388602,0.0001112392,0.4204283,0.0004796063],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001955559,0.00006536337,0.9886754,0.001188134,0.0002274631,0.0004728689,0.0002357242,0.0009133019,0.006266204],"genre_scores_gemma":[0.9681695,0.00005324002,0.02915149,0.001534487,0.0001095221,0.00007393657,0.0008129335,0.0000172607,0.0000775986],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.966214,"threshold_uncertainty_score":0.7819487,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03381081355604557,"score_gpt":0.2467317373066045,"score_spread":0.212920923750559,"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."}}