{"id":"W1963709204","doi":"10.1145/1390334.1390482","title":"Semi-supervised spam filtering","year":2008,"lang":"en","type":"article","venue":"","topic":"Spam and Phishing Detection","field":"Computer Science","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Spambot; Artificial intelligence; Forum spam; Spamming; World Wide Web; The Internet","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":[0.00005672814,0.00005022672,0.00004876589,0.00003746631,0.0001129496,0.00004369648,0.0002922346,0.00002497091,0.00008072132],"category_scores_gemma":[0.00001036029,0.00004449917,0.00002813106,0.0001654078,0.00001042721,0.0003287423,0.00008123437,0.00005071062,0.000184444],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001165288,"about_ca_system_score_gemma":0.000012928,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006437228,"about_ca_topic_score_gemma":0.000004002356,"domain_scores_codex":[0.9995451,0.00001175712,0.00006676638,0.0001533067,0.0001072935,0.0001157749],"domain_scores_gemma":[0.9996505,0.00001953383,0.00001164827,0.0002572702,0.00001718566,0.0000438038],"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":[0.00004360539,0.0004438278,0.0458037,0.00009059427,0.0001142733,0.0005301894,0.01982212,0.002525825,0.3410886,0.09409562,0.1472873,0.3481544],"study_design_scores_gemma":[0.0009229598,0.0002571486,0.04126386,0.0000273632,0.000004404177,0.0005688812,0.00003626147,0.6599222,0.2140897,0.003618815,0.07853611,0.0007522358],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2038751,0.00002499383,0.7672837,0.0003843051,0.0005176037,0.00003762398,1.305039e-7,0.0004362546,0.02744026],"genre_scores_gemma":[0.9698746,0.000009924587,0.02798094,0.0003823383,0.00008771839,0.00000273129,2.885603e-7,0.000003308319,0.00165816],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7659995,"threshold_uncertainty_score":0.2370716,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02946477153280146,"score_gpt":0.2110597670854167,"score_spread":0.1815949955526152,"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."}}