{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003040125,0.00006055458,0.00008305532,0.00006111156,0.0000591584,0.00005237882,0.002053696,0.00003030898,0.00001605004],"category_scores_gemma":[0.000900122,0.00005249683,0.00001480957,0.0002948611,0.00006907939,0.0004056346,0.001305606,0.00007019547,0.0000388772],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001449722,"about_ca_system_score_gemma":0.00006740474,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003095455,"about_ca_topic_score_gemma":0.000003091289,"domain_scores_codex":[0.9990436,0.00001882779,0.0001413478,0.0003208072,0.0002950469,0.0001803947],"domain_scores_gemma":[0.9973601,0.0004663205,0.0000389754,0.001943489,0.000135874,0.00005525709],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00001057968,0.0001565618,0.254205,0.0001882076,0.0001135142,0.000100762,0.00171268,0.00008809211,0.01091699,0.01241835,0.01728178,0.7028075],"study_design_scores_gemma":[0.001150878,0.0008753486,0.1825627,0.0005349507,0.00001792826,0.0003103572,0.00007797155,0.3302352,0.4700526,0.01071921,0.002314406,0.001148439],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02761817,0.0001778061,0.9695421,0.0000687022,0.0003374024,0.00006271717,0.000001615251,0.000454772,0.001736741],"genre_scores_gemma":[0.6031759,0.000002312368,0.3951938,0.00002101621,0.000120502,0.000001685662,0.000001734005,0.000006788501,0.001476211],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7016591,"threshold_uncertainty_score":0.3816312,"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."}}