{"id":"W3176244545","doi":"10.48550/arxiv.2001.01424","title":"Cross-Dataset Design Discussion Mining","year":2020,"lang":"en","type":"preprint","venue":"Figshare","topic":"Data Mining Algorithms and Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Victoria","funders":"","keywords":"Computer science; Artifact (error); Code refactoring; Relevance (law); Context (archaeology); Documentation; Task (project management); Software; Data mining; Machine learning; Software engineering; Artificial intelligence; Data science; Engineering; Systems engineering; Programming language","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","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.00007574895,0.0002382912,0.0002067502,0.00004971283,0.0001699288,0.001113632,0.002787263,0.0001710474,0.01196898],"category_scores_gemma":[0.0004408999,0.0001776589,0.00007233763,0.000208204,0.000007017759,0.0003149841,0.005579506,0.0004019141,0.003563193],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003135869,"about_ca_system_score_gemma":0.0002065983,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004970782,"about_ca_topic_score_gemma":5.012772e-7,"domain_scores_codex":[0.9983003,0.00004679621,0.0002462007,0.0008951495,0.0002653588,0.0002462556],"domain_scores_gemma":[0.9980668,0.0001112268,0.0001970462,0.001386606,0.00006290984,0.0001754093],"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":[6.607154e-7,0.00001410956,0.00000143302,0.0000801995,0.000008130683,0.00001859443,0.0001446329,0.0002647078,0.00000596802,0.00002897604,0.9514848,0.04794779],"study_design_scores_gemma":[0.0001017113,0.00002385892,0.0004139053,0.001386308,0.000005867813,0.000006921249,0.000007280025,0.2894597,0.0002824105,0.0006186209,0.7072647,0.0004286915],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.000001558224,0.00008008913,0.111693,0.001560843,0.0001273501,0.0002974815,0.8856943,0.0003091985,0.0002361684],"genre_scores_gemma":[0.0003157864,0.000002046585,0.2268637,0.0003998179,0.0002303864,0.000549668,0.7714381,0.00002058009,0.0001799451],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.289195,"threshold_uncertainty_score":0.9999233,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1440692337339377,"score_gpt":0.346395833574453,"score_spread":0.2023265998405153,"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."}}