{"id":"W3209931987","doi":"10.5281/zenodo.2583978","title":"The Software Heritage Graph Dataset","year":2019,"lang":"en","type":"dataset","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Scientific Computing and Data Management","field":"Decision Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Graph; Software; Theoretical computer science; Programming language","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":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","sts","scholarly_communication","open_science","insufficient_payload"],"consensus_categories":["open_science","insufficient_payload"],"category_scores_codex":[0.0092655,0.0002783828,0.0002991275,0.0005407961,0.005917274,0.01058374,0.01106699,0.0001202769,0.02022038],"category_scores_gemma":[0.01198618,0.0001972045,0.0001265488,0.001648378,0.0004007768,0.0003553981,0.01083579,0.0005929789,0.1319167],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009544893,"about_ca_system_score_gemma":0.000009425767,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004080021,"about_ca_topic_score_gemma":0.00000418701,"domain_scores_codex":[0.9937118,0.001032136,0.0007274906,0.001429414,0.002490846,0.0006083614],"domain_scores_gemma":[0.9925885,0.0005825023,0.0004609938,0.005358765,0.0007884422,0.0002208215],"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":[0.00002115427,0.00005271766,2.237422e-7,0.00002304823,0.00003293897,0.00001469462,0.00003778726,0.00003422954,0.000001948314,0.0001013441,0.941182,0.05849787],"study_design_scores_gemma":[0.000195082,0.00008581186,0.00004196402,0.00002655605,0.0000216988,0.00003232112,0.0002809078,0.00007154988,0.000001165493,0.0004851887,0.9985181,0.0002396329],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.00001324726,0.0001331397,0.001540161,0.0006449403,0.001362848,0.0005236306,0.9924492,0.0002533882,0.003079501],"genre_scores_gemma":[0.00004789007,0.0001332422,0.00009895255,0.0002863335,0.0002022341,3.887327e-8,0.9955523,0.0004696513,0.003209336],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.1116963,"threshold_uncertainty_score":0.9971644,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1409801775569736,"score_gpt":0.3463128073076928,"score_spread":0.2053326297507192,"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."}}