{"id":"W2572737325","doi":"10.1109/bibm.2016.7822641","title":"Mining sequential patterns from uncertain big DNA in the spark framework","year":2016,"lang":"en","type":"article","venue":"","topic":"Data Mining Algorithms and Applications","field":"Computer Science","cited_by":23,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Manitoba","funders":"","keywords":"Big data; Computer science; Scalability; Data mining; SPARK (programming language); DNA sequencing; Biomedicine; Data science; Bioinformatics; DNA; Database; Biology","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001497736,0.001270416,0.001829925,0.00243977,0.00119228,0.001285153,0.002638057,0.000859957,0.0006049841],"category_scores_gemma":[0.004655768,0.0006635737,0.001737635,0.003117288,0.0006772449,0.001915225,0.001385806,0.0009715544,0.0004934707],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005359966,"about_ca_system_score_gemma":0.002879457,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005189742,"about_ca_topic_score_gemma":0.007074164,"domain_scores_codex":[0.9982881,0.0001635216,0.0001836066,0.0005299845,0.0006532877,0.0001813881],"domain_scores_gemma":[0.9979594,0.0007219017,0.0002964142,0.0003369593,0.0004536955,0.0002316022],"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":[0.001957249,0.0006583406,0.02164879,0.001242859,0.0007113263,0.003263752,0.0008402079,0.4728008,0.0431677,0.02112135,0.02290208,0.4096856],"study_design_scores_gemma":[0.00009668335,0.00008420416,0.001123572,0.00001167546,0.00003854199,0.0004773586,0.0001478696,0.9658762,0.005340761,0.0231248,0.003652726,0.00002563609],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05458055,0.0006752656,0.9327258,0.0004250396,0.0001009375,0.0002719949,0.002277088,0.008262258,0.0006810148],"genre_scores_gemma":[0.197236,0.0003951847,0.7940035,0.0001544779,0.00009119872,0.0003304929,0.006865212,0.0002311652,0.0006927967],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005189742,"threshold_uncertainty_score":0.01031905,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0452226995074654,"score_gpt":0.2825070435540311,"score_spread":0.2372843440465657,"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."}}