{"id":"W4393875236","doi":"10.5281/zenodo.579547","title":"Rediscovery Datasets: Connecting Duplicate Reports Of Libreoffice And Gentoo","year":2017,"lang":"en","type":"dataset","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Computational and Text Analysis Methods","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University","funders":"","keywords":"Computer science; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002057456,0.001738684,0.001130553,0.01233573,0.001303835,0.00202831,0.003164685,0.002421135,0.00901198],"category_scores_gemma":[0.01542893,0.0005964965,0.001375544,0.01020748,0.0007941634,0.001608045,0.003617049,0.001558895,0.01177361],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001761741,"about_ca_system_score_gemma":0.003770089,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03218063,"about_ca_topic_score_gemma":0.04619688,"domain_scores_codex":[0.9967141,0.0004832541,0.0004852667,0.0009071869,0.001086476,0.0003237667],"domain_scores_gemma":[0.9925596,0.002133588,0.001029688,0.001873069,0.001875508,0.0005286297],"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":[0.0002712001,0.0001440362,0.01374314,0.001793628,0.0001576912,0.0003389578,0.0003596772,0.001331538,0.00108975,0.0024463,0.9568853,0.02143884],"study_design_scores_gemma":[0.0003626749,0.00007716338,0.03648151,0.0005972402,0.000150708,0.0006897709,0.0006921896,0.004699416,0.00319636,0.004102011,0.9488336,0.0001172382],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.006244672,0.0004674298,0.0007875634,0.0002432403,0.00006030211,0.00008959079,0.9883462,0.001999343,0.001761669],"genre_scores_gemma":[0.002556213,0.00009149081,0.001395685,0.00004826826,0.00001024471,0.0001104683,0.995181,0.0001189434,0.0004876612],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.03218063,"threshold_uncertainty_score":0.06398666,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06451699705648399,"score_gpt":0.3561331711343312,"score_spread":0.2916161740778472,"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."}}