{"id":"W4230896626","doi":"10.1109/msr.2015.54","title":"Mining StackOverflow to Filter Out Off-Topic IRC Discussion","year":2015,"lang":"en","type":"article","venue":"","topic":"Spam and Phishing Detection","field":"Computer Science","cited_by":31,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"Alberta Innovates - Technology Futures","keywords":"Computer science; Filter (signal processing); Exploit; Set (abstract data type); The Internet; World Wide Web; Stack (abstract data type); Open source; Information retrieval; Online discussion; Data science; Programming language; Computer security","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0002106296,0.00007697674,0.00007456806,0.00006353667,0.00005206926,0.00016918,0.0003629051,0.00003789775,0.00003368184],"category_scores_gemma":[0.00006312624,0.00004719448,0.00002915003,0.0001518275,0.000005740694,0.000421217,0.0001993262,0.00005320562,0.0003574743],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004135184,"about_ca_system_score_gemma":0.00003096589,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002791054,"about_ca_topic_score_gemma":0.00008580092,"domain_scores_codex":[0.9992175,0.00003097946,0.0001051712,0.0002432373,0.0002341616,0.0001690038],"domain_scores_gemma":[0.9993775,0.0000203573,0.00002233841,0.0003599235,0.00004611804,0.0001737568],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0000360549,0.00008800597,0.008362553,0.00001187095,0.00001775682,0.0000298736,0.01924505,0.0004903009,0.002281733,0.007461065,0.2887712,0.6732045],"study_design_scores_gemma":[0.001658331,0.001067185,0.0246397,0.0001068276,0.00001396094,0.00002952936,0.0007742437,0.1361848,0.02304833,0.01153476,0.7998915,0.001050813],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1427945,0.00002206887,0.8252775,0.007027457,0.002919423,0.0001322033,6.360359e-7,0.0003272503,0.02149899],"genre_scores_gemma":[0.924842,8.563482e-7,0.05425772,0.001507601,0.0002734563,0.000008443712,0.000001106232,0.000007664194,0.01910117],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7820475,"threshold_uncertainty_score":0.4594728,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04908666673615086,"score_gpt":0.2695334409950015,"score_spread":0.2204467742588506,"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."}}