{"id":"W4387430331","doi":"10.1007/978-3-031-45275-8_36","title":"Unsupervised Graph Neural Networks for Source Code Similarity Detection","year":2023,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Software Engineering Research","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Ottawa; IBM (Canada); Polytechnique Montréal","funders":"","keywords":"Computer science; Source code; Inference; Similarity (geometry); Artificial intelligence; Graph; Unsupervised learning; Encoder; Artificial neural network; Code (set theory); Pattern recognition (psychology); Context (archaeology); Data mining; Theoretical computer science; Programming language","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":[0.0003707939,0.000704836,0.0006481369,0.001943514,0.0003537791,0.0008090945,0.001457697,0.00110402,0.002658623],"category_scores_gemma":[0.001989046,0.0003999664,0.000646719,0.001947684,0.0003389049,0.001255364,0.0008333268,0.001094641,0.001491609],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006875782,"about_ca_system_score_gemma":0.0005260665,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006426096,"about_ca_topic_score_gemma":0.01346526,"domain_scores_codex":[0.9996295,0.00007181372,0.00001814489,0.0001290458,0.0001148146,0.00003678862],"domain_scores_gemma":[0.9990777,0.0004335178,0.00009804583,0.0001477637,0.0002144328,0.00002867361],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001668542,0.0001515932,0.001668265,0.0001407439,0.00009527968,0.00007184355,0.00004363866,0.1351614,0.01652239,0.005799658,0.009300351,0.830878],"study_design_scores_gemma":[0.00000386132,0.00001836681,0.0005670817,0.000009451366,0.00001245294,0.00003276549,0.00001199729,0.989286,0.00341884,0.005645081,0.0009882919,0.000005722192],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04038147,0.001813475,0.9473708,0.0002751992,0.0001309594,0.00008394791,0.0007744154,0.005522432,0.003647296],"genre_scores_gemma":[0.4454909,0.001143797,0.5327174,0.0001993977,0.0001681272,0.0001590171,0.003246643,0.0007222469,0.01615247],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006426096,"threshold_uncertainty_score":0.01277739,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02761385660487386,"score_gpt":0.2597452961992475,"score_spread":0.2321314395943737,"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."}}