{"id":"W3195417950","doi":"10.1016/j.iot.2021.100444","title":"Automated detection of unstructured context-dependent sensitive information using deep learning","year":2021,"lang":"en","type":"article","venue":"Internet of Things","topic":"Digital and Cyber Forensics","field":"Computer Science","cited_by":15,"is_retracted":false,"has_abstract":false,"ca_institutions":"National Research Council Canada; University of Windsor; University of Victoria","funders":"","keywords":"Context (archaeology); Computer science; Unstructured data; Deep learning; Artificial intelligence; Data science; Machine learning; Information retrieval; Data mining; Big data; Archaeology; Geography","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":[0.0001248592,0.00007731799,0.0001500762,0.00006911267,0.00001925111,0.00008719591,0.0001642463,0.00005063794,0.000002971154],"category_scores_gemma":[0.0001041247,0.00007480702,0.00007214045,0.000169793,0.00003889333,0.001443558,0.0001970338,0.0001035219,0.000005121343],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003856303,"about_ca_system_score_gemma":0.00002855328,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000289191,"about_ca_topic_score_gemma":0.0000124719,"domain_scores_codex":[0.9992468,0.00003954996,0.0002777632,0.0001086078,0.0002284402,0.00009882095],"domain_scores_gemma":[0.9992354,0.0000331079,0.0002392112,0.0001319913,0.000333216,0.00002708726],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00003797947,0.00003673786,0.0002016923,0.00009964743,0.0001882176,0.00002554703,0.02717529,0.003955361,0.08333647,0.01123258,0.00001966471,0.8736908],"study_design_scores_gemma":[0.0001383304,0.00004076208,0.0002334818,0.00005361496,0.000007234516,0.00007030823,0.0003496879,0.5041052,0.4941928,0.0005813981,0.0001666847,0.00006052901],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5197887,0.00002584948,0.4786194,0.00001638146,0.0002879565,0.00003429436,8.44904e-7,0.0001229922,0.0011036],"genre_scores_gemma":[0.995217,0.000001135889,0.004615049,0.00009528033,0.000006368849,3.473364e-7,0.000004415676,0.000003328598,0.00005705941],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8736303,"threshold_uncertainty_score":0.3050542,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.006587824688594163,"score_gpt":0.2080202807576186,"score_spread":0.2014324560690244,"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."}}