{"id":"W2162990323","doi":"10.1109/pst.2008.30","title":"Utility of Knowledge Extracted from Unsanitized Data when Applied to Sanitized Data","year":2008,"lang":"en","type":"article","venue":"","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Calgary","funders":"","keywords":"Computer science; Measure (data warehouse); Data mining; Set (abstract data type); Knowledge extraction; Data science; Data set; Data modeling; Information retrieval; Artificial intelligence; Database","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":["metaresearch","metaepi_narrow","open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.0008086686,0.0002978813,0.0005556236,0.0001977476,0.0001578327,0.00009307907,0.1461885,0.0001936519,0.0005736853],"category_scores_gemma":[0.01781847,0.0002778279,0.00003264737,0.0008083261,0.0002847258,0.001576049,0.3886357,0.0003399639,0.0003078087],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002597178,"about_ca_system_score_gemma":0.0002927671,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008588824,"about_ca_topic_score_gemma":0.000383726,"domain_scores_codex":[0.9963172,0.0001299814,0.0006883765,0.001851741,0.0005130655,0.0004996525],"domain_scores_gemma":[0.9315978,0.0008342102,0.0001937073,0.0670555,0.0001545997,0.0001642255],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00005427615,0.0002962573,0.0005678922,0.00001547499,0.00006977103,0.00002029138,0.0001895385,1.097824e-7,0.00594474,0.001429066,0.9576865,0.0337261],"study_design_scores_gemma":[0.004774191,0.0001063234,0.04193673,0.0001290155,0.00008654757,0.00004364295,0.0001910076,0.5291042,0.05169903,0.2403004,0.1298598,0.001769099],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02168967,0.0002612617,0.9384261,0.004838716,0.0004119723,0.0007548964,0.003554206,0.002534489,0.02752867],"genre_scores_gemma":[0.3575714,0.0000370975,0.6404502,0.0001756933,0.0000435494,0.000009605909,0.001615517,0.00001644126,0.00008049842],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.8278267,"threshold_uncertainty_score":0.9999674,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1864789450298967,"score_gpt":0.322151514453824,"score_spread":0.1356725694239273,"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."}}