{"id":"W4417086205","doi":"10.48550/arxiv.2505.06107","title":"Differentiating Emigration from Return Migration of Scholars Using Name-Based Nationality Detection Models","year":2025,"lang":"en","type":"preprint","venue":"ArXiv.org","topic":"Authorship Attribution and Profiling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Nationality; Emigration; Country of origin; Proxy (statistics); Immigration; TRACE (psycholinguistics)","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["bibliometrics"],"consensus_categories":[],"category_scores_codex":[0.0055059,0.0009158961,0.0008004421,0.004098384,0.001032475,0.002589726,0.001373967,0.001405797,0.001281048],"category_scores_gemma":[0.01306556,0.0002557469,0.001366614,0.002129588,0.0007928952,0.003074104,0.001734883,0.001594892,0.001299639],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009066245,"about_ca_system_score_gemma":0.001162371,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01043866,"about_ca_topic_score_gemma":0.01160373,"domain_scores_codex":[0.9986136,0.0005066002,0.0001329028,0.0004086296,0.0001276824,0.0002106172],"domain_scores_gemma":[0.9932161,0.003317737,0.001232313,0.0007654222,0.001030137,0.0004383018],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005543734,0.0006444098,0.7631789,0.0002122134,0.0002754725,0.0005865527,0.001561088,0.0629378,0.001683134,0.003407004,0.01081594,0.1541431],"study_design_scores_gemma":[0.00004385613,0.0001655629,0.0809942,0.0001587497,0.0001527888,0.0004519598,0.001873072,0.8977302,0.003113329,0.008602877,0.006643817,0.0000695206],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9333901,0.0008519666,0.05828356,0.001375842,0.0002498691,0.000159152,0.001574768,0.0009914042,0.003123343],"genre_scores_gemma":[0.9751768,0.0002310283,0.01887785,0.0002057664,0.00009539694,0.0000783421,0.002956662,0.00006623094,0.002311924],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9959016,"threshold_uncertainty_score":0.02911836,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1163688944268923,"score_gpt":0.3189118797857016,"score_spread":0.2025429853588092,"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."}}