{"id":"W4230052151","doi":"10.32920/ryerson.14641395.v1","title":"Creating stochastic text data to solve privacy issues in social networking","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Spam and Phishing Detection","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University","funders":"","keywords":"Computer science; Social media; Recommender system; Weibull distribution; Data science; Information privacy; Work (physics); World Wide Web; Data modeling; Information retrieval; Internet privacy; Data mining; Database; Engineering","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":[0.01996808,0.0008206493,0.0007376282,0.002369696,0.001635073,0.002419836,0.002231252,0.002575983,0.002209647],"category_scores_gemma":[0.09431215,0.0009030367,0.001380313,0.001958095,0.002292013,0.004719525,0.002839281,0.002822583,0.0009864287],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001638951,"about_ca_system_score_gemma":0.001605091,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002593354,"about_ca_topic_score_gemma":0.002742433,"domain_scores_codex":[0.9809652,0.0116828,0.001068601,0.002489688,0.003378615,0.0004150576],"domain_scores_gemma":[0.9066941,0.06715962,0.00412102,0.01681571,0.004499299,0.0007102023],"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":[0.001177688,0.001222702,0.05898234,0.000620203,0.0004374231,0.001517969,0.003081784,0.586745,0.01950345,0.1320582,0.01571022,0.1789431],"study_design_scores_gemma":[0.00006930772,0.00009012119,0.001791878,0.00002363777,0.00001662234,0.0001944462,0.0001826848,0.9178526,0.007945985,0.06757235,0.00422729,0.00003299932],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1424034,0.0001538445,0.8454026,0.001926307,0.0002459972,0.0006976849,0.004354613,0.002692552,0.002122969],"genre_scores_gemma":[0.5942839,0.0001231551,0.3953347,0.000607801,0.0001466415,0.0009065017,0.006594586,0.0003201567,0.001682584],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01996808,"threshold_uncertainty_score":0.1056026,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1028478848182675,"score_gpt":0.3411414565524468,"score_spread":0.2382935717341793,"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."}}