{"id":"W2953003320","doi":"10.48550/arxiv.1812.10176","title":"A Variability-Aware Design Approach to the Data Analysis Modeling Process","year":2018,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Advanced Software Engineering Methodologies","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Flexibility (engineering); Automation; Process (computing); Data science; Software; Systems engineering; Software engineering; Data mining; Engineering","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","open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.002929684,0.0003900673,0.0005190484,0.0004016698,0.0002369669,0.0001808761,0.01027286,0.0002532128,0.00000315919],"category_scores_gemma":[0.00117937,0.0003441997,0.0001651678,0.002825686,0.00009149466,0.0005468962,0.00846447,0.0006583917,0.00001915114],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001798046,"about_ca_system_score_gemma":0.0002598719,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004371388,"about_ca_topic_score_gemma":0.000006027378,"domain_scores_codex":[0.9960228,0.0007297722,0.0002513391,0.002373825,0.0001694154,0.0004528865],"domain_scores_gemma":[0.9919802,0.0008422045,0.0001669188,0.006517531,0.0003349153,0.0001582376],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001233285,0.00004088019,0.000104761,0.00005633784,0.0003174236,0.000009892246,0.0004558041,0.9952162,5.498783e-7,0.003517823,0.00005403945,0.0002139362],"study_design_scores_gemma":[0.0000649821,0.00001645402,0.0000299059,0.00002049178,0.0003263154,0.000001385324,0.00007170626,0.9323966,0.000008052703,0.06666627,0.00002233987,0.0003755254],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002036459,0.00002308814,0.9960957,0.00006350914,0.0003217901,0.0005691252,0.00003604981,0.0007168261,0.0001374956],"genre_scores_gemma":[0.5252614,0.00001191201,0.4745154,0.00003929522,0.00006518474,0.000004007914,0.00002827826,0.00001484417,0.00005965231],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5232249,"threshold_uncertainty_score":0.999901,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3436730915888954,"score_gpt":0.2658874829567656,"score_spread":0.07778560863212985,"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."}}