{"id":"W1979614540","doi":"","title":"Proceedings of the 18th ACM SIGKDD international conference on Knowledge discovery and data mining","year":2012,"lang":"en","type":"article","venue":"","topic":"Data Mining Algorithms and Applications","field":"Computer Science","cited_by":44,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"","keywords":"Computer science; Knowledge extraction; Data science; Construct (python library); Big data; Field (mathematics); Web mining; Inference; World Wide Web; Artificial intelligence; Web page; Data mining","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":[0.0001979305,0.000054202,0.00005392332,0.00002340076,0.00005795209,0.000156767,0.002266037,0.00001583506,0.000007742083],"category_scores_gemma":[0.00006782632,0.00003419722,0.000008988945,0.0001160878,0.00005180699,0.001249115,0.002356507,0.00004787098,0.000006339179],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000006459639,"about_ca_system_score_gemma":0.00002403476,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000009510537,"about_ca_topic_score_gemma":0.000002289853,"domain_scores_codex":[0.9994925,0.000003411354,0.00009605941,0.0001956458,0.0001129699,0.00009942683],"domain_scores_gemma":[0.9992654,0.00005509821,0.00005719357,0.0005398369,0.000048568,0.00003395451],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000001521732,0.0001550988,0.01058383,0.00000967562,0.0000160378,3.399862e-8,0.001522885,1.132301e-7,0.001687677,0.8706052,0.02363891,0.09177908],"study_design_scores_gemma":[0.001248226,0.0001808355,0.2692316,0.0005023762,0.00005346271,0.00006121537,0.003680028,0.5375378,0.02019592,0.00842165,0.157908,0.0009788437],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5613441,0.000185114,0.1197486,0.01244895,0.001485979,0.00047725,0.0003115853,0.0001766151,0.3038219],"genre_scores_gemma":[0.9220988,0.00001245111,0.07579423,0.00009244999,0.00008243942,0.000006704272,0.000008949632,0.000002828677,0.001901157],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8621835,"threshold_uncertainty_score":0.4210898,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09907508012851189,"score_gpt":0.3369918420858378,"score_spread":0.2379167619573259,"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."}}