{"id":"W2888961423","doi":"","title":"Poster: Fast, Scalable and User-Guided Clone Detection","year":2018,"lang":"en","type":"article","venue":"International Conference on Software Engineering","topic":"Software Engineering Research","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Saskatchewan","funders":"","keywords":"clone (Java method); Computer science; Scalability; Source code; Transformation (genetics); Search engine indexing; Jaccard index; Data mining; Artificial intelligence; Database; Programming language; Pattern recognition (psychology); 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002194179,0.0002007496,0.0001447921,0.0003439455,0.000082078,0.00034293,0.0007524878,0.00008671016,0.00007195058],"category_scores_gemma":[0.0009443548,0.0002126574,0.00003862253,0.0002778732,0.00004639093,0.0005304821,0.0003154412,0.0002563082,0.0001519327],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001254165,"about_ca_system_score_gemma":0.0000444187,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003007388,"about_ca_topic_score_gemma":0.000004465441,"domain_scores_codex":[0.9984819,0.00001638478,0.0002154099,0.0004562454,0.0005001532,0.00032986],"domain_scores_gemma":[0.9988045,0.0002538663,0.00004288738,0.0003726751,0.0003796421,0.0001465059],"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":[0.0002793817,0.0005928944,0.1551576,0.0004316132,0.00103054,0.0004535366,0.004801254,0.01079317,0.1697338,0.269439,0.006101639,0.3811856],"study_design_scores_gemma":[0.001876681,0.0009562416,0.1699794,0.0005904523,0.00001384436,0.0004457686,0.00003568193,0.6832985,0.123506,0.0008648771,0.01687329,0.001559277],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2071797,0.00002202522,0.7900175,0.0002417974,0.001284717,0.0001119838,0.000005428282,0.0007805038,0.0003563712],"genre_scores_gemma":[0.9676412,0.0000154053,0.03149445,0.0000819044,0.000284999,0.00002923662,0.000003644837,0.00002509861,0.0004240451],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7604616,"threshold_uncertainty_score":0.8671915,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0295438719690846,"score_gpt":0.2812903793180453,"score_spread":0.2517465073489608,"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."}}