{"id":"W2955898947","doi":"10.1145/3331184.3331395","title":"Information Retrieval Meets Scalable Text Analytics","year":2019,"lang":"en","type":"article","venue":"","topic":"Advanced Database Systems and Queries","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"SPARK (programming language); Scalability; Computer science; Analytics; Downstream (manufacturing); Pipeline transport; Range (aeronautics); Data science; Database; Programming language; Engineering","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":[0.00987419,0.001805792,0.002348202,0.003664534,0.001874499,0.01049758,0.003784814,0.003044633,0.01353535],"category_scores_gemma":[0.05323259,0.001187885,0.001566299,0.007256145,0.001995982,0.01780027,0.008170595,0.003212599,0.02042309],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001662108,"about_ca_system_score_gemma":0.004198554,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004040479,"about_ca_topic_score_gemma":0.004489094,"domain_scores_codex":[0.9767556,0.004974215,0.002101403,0.004087281,0.01062183,0.001459793],"domain_scores_gemma":[0.966705,0.01192175,0.001660237,0.01134434,0.007098207,0.001270433],"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":[0.001466739,0.0006717368,0.005744459,0.002299693,0.0003708327,0.001312678,0.0009252166,0.0443364,0.02725009,0.1548806,0.2947058,0.4660358],"study_design_scores_gemma":[0.0003340056,0.0001827775,0.001688796,0.0001465864,0.0001057016,0.001249057,0.0008437259,0.4178258,0.01902426,0.445926,0.1125507,0.0001224904],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03692512,0.005466299,0.79768,0.02554858,0.001155833,0.001258895,0.01402504,0.0490176,0.06892262],"genre_scores_gemma":[0.3258457,0.00253514,0.6283544,0.002813006,0.00196658,0.0009350695,0.02076146,0.003202538,0.01358606],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01353535,"threshold_uncertainty_score":0.05222034,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008595608315333266,"score_gpt":0.2252329320424834,"score_spread":0.2166373237271501,"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."}}