{"id":"W3210216704","doi":"10.1109/dsaa53316.2021.9564218","title":"DaskDB: Scalable Data Science with Unified Data Analytics and In Situ Query Processing","year":2021,"lang":"en","type":"article","venue":"","topic":"Advanced Database Systems and Queries","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of New Brunswick","funders":"","keywords":"Computer science; Python (programming language); Scalability; SQL; Relational database management system; Analytics; Database; Relational database; Benchmark (surveying); Data analysis; Usability; Data mining; Programming language; Operating system","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":[],"consensus_categories":[],"category_scores_codex":[0.0007447887,0.000111227,0.0001686727,0.0001018824,0.0001745879,0.0003341581,0.001640265,0.00002096312,0.000004047088],"category_scores_gemma":[0.000153775,0.00008399823,0.000003051015,0.001596718,0.0002180675,0.007936933,0.004536636,0.0001026593,0.000002950661],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002223346,"about_ca_system_score_gemma":0.0008859173,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001256312,"about_ca_topic_score_gemma":0.002802295,"domain_scores_codex":[0.9981117,0.00002714368,0.0001989893,0.001022732,0.0003440165,0.0002953715],"domain_scores_gemma":[0.9965385,0.00004665309,0.00006422073,0.003117507,0.0001324847,0.0001006297],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001097138,0.0008619557,0.0605062,0.001403567,0.00008271651,0.002826122,0.002881689,0.002512186,0.06871637,0.5277862,0.004515533,0.3277978],"study_design_scores_gemma":[0.0008007602,0.00004481644,0.007878856,0.0004705299,0.00001499001,0.0003473835,0.001560689,0.9387485,0.01150166,0.000433175,0.03753099,0.0006676348],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01360247,0.0005177841,0.9815027,0.0006167392,0.00005993346,0.00008529909,0.00005005191,0.00007393982,0.003491082],"genre_scores_gemma":[0.4740496,0.00006006153,0.5252787,0.0001703752,0.0000247895,0.000002262738,0.0001311895,0.000006583454,0.0002764166],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9362363,"threshold_uncertainty_score":0.5754085,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07135703472201405,"score_gpt":0.3065453377071228,"score_spread":0.2351883029851087,"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."}}