{"id":"W2901077691","doi":"10.1109/icsme.2018.00009","title":"Threats of Aggregating Software Repository Data","year":2018,"lang":"en","type":"article","venue":"","topic":"Software Engineering Research","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Artifact (error); Software; Open source software; Set (abstract data type); Open source; Database; Software engineering; Operating system; Programming language","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.03967187,0.0008857741,0.001061478,0.01132341,0.004036582,0.004836616,0.002068926,0.003312254,0.0006920422],"category_scores_gemma":[0.1980378,0.0006258356,0.001121819,0.0139995,0.003914379,0.006024033,0.007237236,0.005066395,0.0004762852],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001814866,"about_ca_system_score_gemma":0.003239421,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004142103,"about_ca_topic_score_gemma":0.003001594,"domain_scores_codex":[0.8780958,0.04194081,0.01069579,0.006594007,0.05940408,0.003269506],"domain_scores_gemma":[0.519366,0.2822726,0.06702301,0.08507066,0.04243131,0.003836335],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"not_applicable","study_design_scores_codex":[0.001901395,0.001361134,0.69628,0.001620448,0.0008161717,0.004650291,0.02421918,0.023717,0.009133096,0.04404645,0.0260273,0.1662275],"study_design_scores_gemma":[0.0004012063,0.001816409,0.4616053,0.003200921,0.001293926,0.021865,0.02776928,0.1509248,0.06593815,0.08425111,0.1801717,0.0007621341],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.939335,0.0006581644,0.0323282,0.004633884,0.0002323453,0.0007453593,0.006082854,0.001018278,0.01496587],"genre_scores_gemma":[0.9689128,0.0002393521,0.02145612,0.0009218765,0.0001052923,0.000419578,0.006960338,0.0001838056,0.000800642],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.03967187,"threshold_uncertainty_score":0.2098074,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05429274992367693,"score_gpt":0.3142213925941972,"score_spread":0.2599286426705202,"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."}}