{"id":"W3186295254","doi":"10.22215/etd/2021-14525","title":"A Workload-Driven Framework for NoSQL Data Modeling and Partitioning","year":2021,"lang":"en","type":"dissertation","venue":"","topic":"Graph Theory and Algorithms","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University","funders":"","keywords":"NoSQL; Computer science; Scalability; Heuristics; Database; Big data; Schema (genetic algorithms); Data mining; Workload; Distributed computing; Information retrieval","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":[0.0002197299,0.000164371,0.0002076838,0.00007147689,0.0002275583,0.0004176719,0.0008753469,0.0001930858,0.00003326282],"category_scores_gemma":[0.0000973267,0.0001620104,0.00005465382,0.0001752608,0.00001007272,0.0004496211,0.0002146297,0.0002226376,0.000004420524],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000005725374,"about_ca_system_score_gemma":0.00007105455,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000005804236,"about_ca_topic_score_gemma":0.00002725473,"domain_scores_codex":[0.9987439,0.00003332055,0.0002039426,0.0006686174,0.0001397157,0.000210501],"domain_scores_gemma":[0.9987417,0.0001820816,0.00007866447,0.0008295143,0.00009976642,0.00006820884],"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":[0.00001002082,0.000028899,0.000009814463,0.0001123159,0.00005929234,0.0000072579,0.0009964891,0.001268879,0.00002467558,0.9449375,0.00017455,0.05237031],"study_design_scores_gemma":[0.00007537795,0.00001261082,0.000006667959,0.0004465075,0.000020242,0.000002315811,0.0003209283,0.7290305,0.00008583341,0.2696619,0.0001549531,0.000182194],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.002140161,0.001516721,0.9943497,0.0001575865,0.0007101637,0.0001632966,0.00001883658,0.00009868741,0.0008448515],"genre_scores_gemma":[0.05707741,0.0002843032,0.9383973,0.0003333305,0.0002460916,0.00007134696,0.001437051,0.00002585553,0.002127271],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.7277616,"threshold_uncertainty_score":0.6606591,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05358533181343143,"score_gpt":0.3190328447488879,"score_spread":0.2654475129354565,"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."}}