{"id":"W6891414117","doi":"10.3929/ethz-a-010137112","title":"Comparing ICP variants on real-world data sets: Open-source library and experimental protocol","year":2013,"lang":"en","type":"article","venue":"reroDoc Digital Library","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Fonds Québécois de la Recherche sur la Nature et les Technologies","keywords":"Protocol (science); Data collection; Identification (biology); Matching (statistics); Feature (linguistics)","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication","open_science"],"consensus_categories":["scholarly_communication"],"category_scores_codex":[0.00006890303,0.0003119789,0.0003251654,0.0001868424,0.0001605258,0.008765427,0.004883982,0.00005909543,0.00008198207],"category_scores_gemma":[0.00002075671,0.0002590926,0.00003004574,0.0005689599,0.0001199439,0.05074877,0.01192884,0.0002720637,0.0001052375],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001190863,"about_ca_system_score_gemma":0.00009439129,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003173044,"about_ca_topic_score_gemma":3.583584e-7,"domain_scores_codex":[0.9979218,0.00007112003,0.0003569779,0.0009769218,0.0002777636,0.0003954098],"domain_scores_gemma":[0.9977958,0.0001225652,0.0001811465,0.001652095,0.000008792789,0.0002396294],"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":[0.0003500254,0.001392919,0.03628441,0.0002891282,0.00009838531,0.0002718653,0.0004644499,0.000003812409,0.001133327,0.2203342,0.4028399,0.3365377],"study_design_scores_gemma":[0.005404822,0.002236048,0.02458813,0.003070295,0.00001954585,0.0003518398,0.0002126782,0.0648131,0.1916844,0.1323927,0.5701562,0.005070208],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.07150503,0.001109099,0.00928252,0.01432633,0.0004694146,0.1450613,0.0005293149,0.02052513,0.7371919],"genre_scores_gemma":[0.6421097,0.000009054868,0.3312973,0.003499936,0.0002240832,0.01202544,0.0005320128,0.0001833173,0.0101191],"genre_candidate":"other","genre_consensus":null,"teacher_disagreement_score":0.7270728,"threshold_uncertainty_score":0.9999861,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04447859576449631,"score_gpt":0.3107546281914985,"score_spread":0.2662760324270022,"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."}}