{"id":"W4362603997","doi":"10.1016/j.infsof.2023.107218","title":"Dev2vec: Representing domain expertise of developers in an embedding space","year":2023,"lang":"en","type":"article","venue":"Information and Software Technology","topic":"Software Engineering Research","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":false,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Embedding; Computer science; Domain (mathematical analysis); Space (punctuation); Software engineering; Data science; Software; World Wide Web; Information retrieval; Knowledge management; Artificial intelligence; Programming language; Mathematics","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":[0.000413963,0.00007840537,0.0001251908,0.001270924,0.00005342474,0.00005588178,0.0004472905,0.0001261053,0.000002762517],"category_scores_gemma":[0.001483398,0.00008072538,0.00001288968,0.001773429,0.00006215633,0.001222632,0.0003812358,0.0001536601,0.00002305562],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004188062,"about_ca_system_score_gemma":0.0000503792,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001894726,"about_ca_topic_score_gemma":0.000007168725,"domain_scores_codex":[0.9991282,0.00001712084,0.0002818777,0.0001385439,0.0001760908,0.0002581665],"domain_scores_gemma":[0.999297,0.0001645939,0.00006247767,0.0003456094,0.00008762449,0.00004266432],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.000007289666,0.00001760582,0.4037059,0.0001153805,0.00001039489,0.00002985819,0.01397977,0.002518773,0.0005659624,0.02024493,0.0002318385,0.5585724],"study_design_scores_gemma":[0.003275821,0.0003921014,0.5020711,0.0004728485,0.000004221763,0.000236087,0.0111394,0.4275895,0.02018693,0.02207361,0.01116728,0.001391137],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7111467,0.00002888395,0.2871597,0.0004144848,0.00005586106,0.00009187939,6.305441e-7,0.001073316,0.00002854492],"genre_scores_gemma":[0.8886729,0.00002039536,0.1112258,0.00002726192,0.000003563989,0.00002989909,0.000004513259,0.000004888736,0.00001073758],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5571812,"threshold_uncertainty_score":0.3291886,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0143342948607931,"score_gpt":0.2854876674304545,"score_spread":0.2711533725696614,"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."}}