{"id":"W4221058968","doi":"10.32604/iasc.2022.027349","title":"Improve Representation for Cross-Language Clone Detection by Pretrain Using Tree Autoencoder","year":2022,"lang":"en","type":"article","venue":"Intelligent Automation & Soft Computing","topic":"Software Engineering Research","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Computer science; Autoencoder; Artificial intelligence; Deep learning; Tree (set theory); Context (archaeology); Node (physics); Encoder; Embedding; Feature learning; Pattern recognition (psychology); Machine learning","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":[0.001013334,0.001487566,0.000882033,0.001013079,0.0003826653,0.000793984,0.001679387,0.0009712041,0.001681465],"category_scores_gemma":[0.004075002,0.0005030432,0.001019365,0.0008186186,0.0005095626,0.002584525,0.001515436,0.001904006,0.001313992],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000827872,"about_ca_system_score_gemma":0.001447842,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01144423,"about_ca_topic_score_gemma":0.01428082,"domain_scores_codex":[0.9992046,0.00009086743,0.00005515232,0.0003430958,0.000198692,0.0001076443],"domain_scores_gemma":[0.9986163,0.000398798,0.000128098,0.0002678028,0.0005338684,0.00005506316],"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":[0.0002440964,0.0002381565,0.01026564,0.0001510236,0.0001565352,0.0003475993,0.0002767442,0.1641021,0.03215772,0.003062346,0.009737155,0.7792608],"study_design_scores_gemma":[0.00001196312,0.00006705496,0.00122148,0.00001178521,0.00002793646,0.00008293531,0.00003612407,0.9817202,0.01322788,0.002051363,0.001523739,0.00001749946],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1264461,0.0006390333,0.8560947,0.0003158698,0.0001668684,0.0001145755,0.0006604143,0.01301628,0.002546113],"genre_scores_gemma":[0.6690307,0.0004171997,0.314391,0.0007700983,0.00006279874,0.0002838992,0.005596604,0.0008354652,0.008612285],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01144423,"threshold_uncertainty_score":0.02275521,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02670070496416517,"score_gpt":0.3361846129183926,"score_spread":0.3094839079542274,"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."}}