{"id":"W3012083778","doi":"10.48550/arxiv.2003.04972","title":"A Comparative Study of Sequence Classification Models for Privacy Policy Coverage Analysis","year":2020,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Privacy, Security, and Data Protection","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Guelph","funders":"","keywords":"Computer science; Jargon; GRASP; Privacy policy; Microdata (statistics); Classifier (UML); Information privacy; Data science; Information retrieval; Internet privacy; Data mining; World Wide Web; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.02821265,0.001956276,0.002688793,0.00649916,0.001596475,0.004560904,0.002979469,0.003007398,0.004782981],"category_scores_gemma":[0.06978641,0.0005764675,0.002411956,0.005624069,0.001547276,0.008675518,0.002195345,0.005907716,0.001694638],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.006021033,"about_ca_system_score_gemma":0.003787139,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02753722,"about_ca_topic_score_gemma":0.01766223,"domain_scores_codex":[0.9876072,0.007958473,0.0006527263,0.001486652,0.001752295,0.0005426172],"domain_scores_gemma":[0.8550468,0.1283832,0.002515565,0.005174471,0.007430581,0.001449355],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.002028757,0.001263008,0.03147622,0.0005748679,0.0006598639,0.0001855521,0.00117127,0.4100696,0.0006160529,0.06949006,0.02218476,0.4602799],"study_design_scores_gemma":[0.00001510914,0.00006327723,0.0007286283,0.00004628662,0.0000294391,0.00002212411,0.00009507211,0.9802991,0.000120683,0.01745855,0.001107641,0.00001404748],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.187218,0.01902442,0.7555083,0.01253138,0.0008393598,0.0006533189,0.002690103,0.002579593,0.01895556],"genre_scores_gemma":[0.8410597,0.0052734,0.1383703,0.001598505,0.0007981836,0.0004554056,0.004941699,0.0004612638,0.007041638],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02821265,"threshold_uncertainty_score":0.1492046,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.322944510737313,"score_gpt":0.3236108289440943,"score_spread":0.0006663182067812579,"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."}}