{"id":"W4401508264","doi":"10.1109/dcoss-iot61029.2024.00037","title":"Privacy-Preserving Multi-Party Keyword-Based Classification of Unstructured Text Data","year":2024,"lang":"en","type":"article","venue":"","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University","funders":"","keywords":"Computer science; Information retrieval; Unstructured data; Keyword search; Natural language processing; World Wide Web; Data mining; Big data","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":[0.0002764003,0.000130808,0.0001371674,0.0002333309,0.00006438228,0.0002810387,0.00400198,0.0000979549,0.000119157],"category_scores_gemma":[0.0002960383,0.0001026973,0.00004313813,0.0007990872,0.00009496926,0.001073667,0.00118025,0.0001401925,0.00006962098],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003337015,"about_ca_system_score_gemma":0.0001364942,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002869578,"about_ca_topic_score_gemma":0.0000189231,"domain_scores_codex":[0.9985592,0.00003881898,0.000334517,0.0006100985,0.0002720024,0.0001853319],"domain_scores_gemma":[0.9966199,0.0001933676,0.00009707652,0.002980239,0.00006797959,0.00004142248],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000006285178,0.0001212924,0.002280293,0.0001871233,0.00004501767,0.000004387643,0.0002689702,0.00006003716,0.05810684,0.553587,0.0454325,0.3399003],"study_design_scores_gemma":[0.0001909629,0.00001988912,0.01060406,0.00004140583,0.000007988056,0.000001473706,0.00006866037,0.9129182,0.03117904,0.006359856,0.03845095,0.0001575262],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.003273818,0.0003657254,0.9822898,0.00980992,0.000352277,0.0002035597,0.00001597723,0.001625298,0.00206365],"genre_scores_gemma":[0.7957194,0.00001368322,0.2034389,0.00006741397,0.0000151163,0.00001179897,0.00004285954,0.000007983608,0.0006828009],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9128581,"threshold_uncertainty_score":0.7436741,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.10229281079065,"score_gpt":0.3338654759186699,"score_spread":0.2315726651280198,"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."}}