{"id":"W4393451681","doi":"10.5281/zenodo.2558378","title":"SeSaMe: A Data Set of Semantically Similar Java Methods","year":2019,"lang":"en","type":"dataset","venue":"Figshare","topic":"Scientific Computing and Data Management","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Java; Computer science; Set (abstract data type); Programming language; Information retrieval; Database","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001796324,0.00221732,0.00122714,0.007871499,0.001485877,0.002401685,0.003309181,0.002808395,0.01280512],"category_scores_gemma":[0.01131157,0.0007809284,0.002473172,0.00849382,0.0007537794,0.002480329,0.003340853,0.003010379,0.01573085],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001686496,"about_ca_system_score_gemma":0.003451291,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01981815,"about_ca_topic_score_gemma":0.03370014,"domain_scores_codex":[0.9966485,0.0004300249,0.0007359681,0.0008010999,0.001046685,0.000337672],"domain_scores_gemma":[0.9942207,0.001710999,0.0006615636,0.001416512,0.001479486,0.0005106897],"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":[0.0006890846,0.000225819,0.005975192,0.004122029,0.0001807346,0.0002859314,0.0002550787,0.00134544,0.002532452,0.004789334,0.9619576,0.01764136],"study_design_scores_gemma":[0.000528149,0.00007138584,0.0121967,0.0005788124,0.00008976362,0.0003749804,0.0002946746,0.001535444,0.002830624,0.003156117,0.9782614,0.00008201195],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.001576626,0.0002048409,0.000704887,0.0001622734,0.00006581915,0.00007497688,0.9950147,0.001095163,0.001100625],"genre_scores_gemma":[0.00166198,0.00009065202,0.001816348,0.00008018535,0.000007632345,0.0002236031,0.9956825,0.0001225072,0.0003146112],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.01981815,"threshold_uncertainty_score":0.04283732,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.6983310733262271,"score_gpt":0.5754483102370438,"score_spread":0.1228827630891833,"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."}}