{"id":"W2160590935","doi":"10.1109/icsm.2009.5306383","title":"An empirical study on the risks of using off-the-shelf techniques for processing mailing list data","year":2009,"lang":"en","type":"article","venue":"","topic":"Software System Performance and Reliability","field":"Computer Science","cited_by":34,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University","funders":"","keywords":"Computer science; Data processing; Software; Data science; Empirical research; World Wide Web; Database; Information retrieval; Data mining","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":[0.08250603,0.0004887534,0.0004186127,0.00238351,0.001500066,0.003179327,0.001528477,0.001920304,0.002803392],"category_scores_gemma":[0.4951151,0.0005171761,0.0006628544,0.002921069,0.002016881,0.006933843,0.001863919,0.00344597,0.000907714],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001371232,"about_ca_system_score_gemma":0.002167273,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002867017,"about_ca_topic_score_gemma":0.002544511,"domain_scores_codex":[0.9039966,0.06608915,0.008330564,0.002904012,0.01734549,0.00133414],"domain_scores_gemma":[0.1125309,0.7949793,0.04462812,0.02133253,0.02441765,0.002111592],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002387814,0.006156144,0.8301643,0.001114366,0.0003079743,0.0007277112,0.02163548,0.002332674,0.00173811,0.002864561,0.003369949,0.1272009],"study_design_scores_gemma":[0.0004267287,0.01346288,0.8373973,0.002235634,0.0008372886,0.00511254,0.06213809,0.04203828,0.009613633,0.005060408,0.02133008,0.0003470081],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9893621,0.0004937425,0.004434687,0.001173294,0.0000296505,0.0002934624,0.0002326891,0.00006997838,0.003910454],"genre_scores_gemma":[0.9925435,0.0003580078,0.005553934,0.0002671902,0.00002821449,0.000163434,0.0002333259,0.00003955949,0.0008129185],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.08250603,"threshold_uncertainty_score":0.4363388,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2279503073911953,"score_gpt":0.4522758268410009,"score_spread":0.2243255194498057,"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."}}