{"id":"W3143356065","doi":"10.1145/3448016.3457330","title":"DataPrep.EDA: Task-Centric Exploratory Data Analysis for Statistical Modeling in Python","year":2021,"lang":"en","type":"article","venue":"","topic":"Data Visualization and Analytics","field":"Computer Science","cited_by":29,"is_retracted":false,"has_abstract":true,"ca_institutions":"Simon Fraser University","funders":"Mitacs; National Science Foundation","keywords":"Python (programming language); Granularity; Exploratory data analysis; Statistical model; Data modeling; Data structure; Data exploration","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":[0.008638716,0.002523031,0.001623288,0.002113222,0.00146784,0.005492316,0.005536598,0.00117536,0.040439],"category_scores_gemma":[0.02744673,0.002534532,0.003838733,0.002203731,0.002095626,0.007097709,0.007858331,0.006860382,0.02514927],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001405901,"about_ca_system_score_gemma":0.006527198,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003981636,"about_ca_topic_score_gemma":0.006312364,"domain_scores_codex":[0.9967346,0.0008962246,0.0004033253,0.0006994821,0.0009729614,0.0002934027],"domain_scores_gemma":[0.9846517,0.007690478,0.0009307973,0.004045897,0.001820922,0.0008601677],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0009284792,0.0003052771,0.00926562,0.0029499,0.0007690107,0.0005696772,0.001747625,0.01453742,0.009834653,0.04940239,0.7622786,0.1474113],"study_design_scores_gemma":[0.0007676783,0.000160408,0.006447949,0.0004831801,0.0001940606,0.0006906532,0.0003558937,0.1864949,0.0322215,0.1984335,0.5731763,0.0005738492],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"software","genre_gemma":"software","genre_scores_codex":[0.001820544,0.0001373111,0.4256861,0.0008892567,0.0002130941,0.0003811364,0.0228361,0.5442628,0.00377371],"genre_scores_gemma":[0.05126217,0.0005702534,0.6777034,0.002766789,0.0002357426,0.003725608,0.04844886,0.2063646,0.008922493],"genre_candidate":"software","genre_consensus":"software","teacher_disagreement_score":0.040439,"threshold_uncertainty_score":0.135282,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1746381258153367,"score_gpt":0.3712001494054153,"score_spread":0.1965620235900786,"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."}}