{"id":"W4417132523","doi":"10.1109/saner-c66551.2025.00015","title":"Assessing Data Augmentation-Induced Bias in Training and Testing of Machine Learning Models","year":2025,"lang":"","type":"article","venue":"","topic":"Software Testing and Debugging Techniques","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Ontario Tech University","funders":"","keywords":"Debugging; Training set; Software; Test data; Test (biology); Training (meteorology); Non-regression testing; Keyword-driven testing; Software testing","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.003134021,0.0002790691,0.0004839441,0.0006393285,0.0002874727,0.000698798,0.001116033,0.0001247695,0.000004570261],"category_scores_gemma":[0.005761156,0.0002925669,0.00002754753,0.001880047,0.00009300768,0.002294536,0.001755529,0.0005391894,4.60019e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005799226,"about_ca_system_score_gemma":0.0005307381,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002600839,"about_ca_topic_score_gemma":0.00005617095,"domain_scores_codex":[0.9970927,0.0003452454,0.0008885851,0.0009435329,0.0003152579,0.0004146834],"domain_scores_gemma":[0.9945691,0.003885567,0.0003883255,0.0009175916,0.0001658244,0.00007354502],"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":[0.000003750016,0.00008575393,0.08075041,0.0001748777,0.00002516277,0.00001478767,0.003435118,0.002638012,0.001053191,0.002331971,0.00008615066,0.9094008],"study_design_scores_gemma":[0.000445365,0.00007442163,0.004398835,0.00195018,0.00002364133,0.00001422719,0.0002577218,0.9633355,0.00077848,0.02847191,0.000007919217,0.0002418581],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1788432,0.0009415351,0.8140272,0.0003085918,0.0001654721,0.0002166188,0.000003075584,0.002068551,0.00342581],"genre_scores_gemma":[0.6492382,0.00002846529,0.3505531,0.00009733364,0.0000118183,0.00000407675,0.000007037152,0.00001052996,0.0000494556],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9606974,"threshold_uncertainty_score":0.9999527,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4312538849540782,"score_gpt":0.4003276714823415,"score_spread":0.03092621347173669,"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."}}