{"id":"W2941695365","doi":"10.48550/arxiv.1904.10403","title":"Optimizing Search API Queries for Twitter Topic Classifiers Using a Maximum Set Coverage Approach","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Classifier (UML); Information retrieval; Set (abstract data type); Precision and recall; Data mining; Training set; Machine learning; Artificial intelligence","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.000271576,0.0003530944,0.0004006599,0.0004189125,0.0002412008,0.0003998134,0.002162758,0.0004618298,0.00001466475],"category_scores_gemma":[0.00002731455,0.0003895709,0.0002560446,0.0004651919,0.0001827982,0.0006227932,0.002220566,0.0005687813,0.00002627414],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000377288,"about_ca_system_score_gemma":0.0002859937,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005676405,"about_ca_topic_score_gemma":0.000002103301,"domain_scores_codex":[0.9977369,0.00009274619,0.0002382346,0.001288407,0.0001360682,0.0005076342],"domain_scores_gemma":[0.997803,0.0001166799,0.0002432714,0.001587607,0.0001565349,0.00009295045],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00007741193,0.0001366713,0.003259726,0.0006109001,0.0002357861,0.00004607229,0.001840316,0.5827487,0.0003423297,0.4072452,0.001259002,0.002197894],"study_design_scores_gemma":[0.0007428511,0.00006523017,0.0002068689,0.00006970618,0.00005418126,0.000004383714,0.00106926,0.9438224,0.0007311825,0.0495354,0.003031715,0.0006668383],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1153954,0.0000546689,0.8806947,0.0002891014,0.0004861408,0.0007344627,0.00001883769,0.0004756567,0.001851014],"genre_scores_gemma":[0.9511432,0.0001033818,0.04528035,0.0001153094,0.00004803307,0.00000643251,0.00003731429,0.00002414267,0.003241838],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8357478,"threshold_uncertainty_score":0.9998556,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2023589357258472,"score_gpt":0.2432152504124623,"score_spread":0.0408563146866151,"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."}}