{"id":"W3048081621","doi":"10.48550/arxiv.2008.02954","title":"Deep Active Learning with Crowdsourcing Data for Privacy Policy Classification","year":2020,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Mobile Crowdsensing and Crowdsourcing","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Crowdsourcing; Computer science; Leverage (statistics); Classifier (UML); Machine learning; Artificial intelligence; Privacy policy; Annotation; Labeled data; Class (philosophy); Active learning (machine learning); Information privacy; World Wide Web; Internet privacy","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":[0.0057925,0.001254596,0.001200003,0.00189228,0.001437306,0.002264668,0.00273733,0.002338323,0.002449049],"category_scores_gemma":[0.01394423,0.0007094577,0.001103712,0.001778392,0.002070735,0.003046126,0.002953761,0.003817804,0.001314698],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003076111,"about_ca_system_score_gemma":0.002854798,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01407806,"about_ca_topic_score_gemma":0.01743833,"domain_scores_codex":[0.9967589,0.001294735,0.0001497615,0.0008653851,0.0006138249,0.0003175009],"domain_scores_gemma":[0.992442,0.004593155,0.0005420722,0.001292122,0.0008466293,0.000284076],"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":[0.000697275,0.0006905647,0.006245873,0.0002411609,0.0001405339,0.0002751857,0.000668278,0.5608294,0.006055328,0.03031497,0.01794241,0.375899],"study_design_scores_gemma":[0.0000172557,0.00001699359,0.000188062,0.00001418926,0.000006964008,0.000008980412,0.0000322061,0.9790151,0.001858929,0.01644766,0.002384948,0.000008819061],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04277154,0.0006192278,0.9415864,0.002699485,0.0002508147,0.0002881097,0.0009891959,0.003883571,0.006911695],"genre_scores_gemma":[0.7264443,0.0003455971,0.2591174,0.001237495,0.0002208919,0.0005555239,0.002236967,0.0002690592,0.009572717],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01407806,"threshold_uncertainty_score":0.03063405,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1448672035770451,"score_gpt":0.2307093802426244,"score_spread":0.08584217666557933,"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."}}