{"id":"W4225003277","doi":"10.1016/j.trc.2022.103668","title":"Transformer-based map-matching model with limited labeled data using transfer-learning approach","year":2022,"lang":"en","type":"article","venue":"Transportation Research Part C Emerging Technologies","topic":"Traffic Prediction and Management Techniques","field":"Engineering","cited_by":43,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University","funders":"","keywords":"Map matching; Computer science; Transformer; Artificial intelligence; Ground truth; Matching (statistics); Data mining; Raw data; Transfer of learning; Pattern recognition (psychology); Mathematics; Global Positioning System; Engineering","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001065725,0.0002634658,0.0002626369,0.0009134943,0.0008330508,0.0000635666,0.001031605,0.0001115489,0.00002197614],"category_scores_gemma":[0.00001224435,0.0002698537,0.0000550984,0.001363025,0.0001760319,0.0004338113,0.00003422965,0.00147479,0.000001660711],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001345968,"about_ca_system_score_gemma":0.00007593861,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006001593,"about_ca_topic_score_gemma":0.00008833387,"domain_scores_codex":[0.9973103,0.0001003884,0.0004009972,0.0005508602,0.0009714951,0.0006660051],"domain_scores_gemma":[0.9990854,0.00006005979,0.00002902878,0.0006880621,0.00008103112,0.0000563643],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00006504986,0.00008953554,0.0001473783,0.0002545494,0.00007978848,0.00001486773,0.0005765294,0.9860327,0.003595387,0.001773004,0.002135237,0.005235936],"study_design_scores_gemma":[0.0006577148,0.0001221381,0.00002678323,0.0000567297,0.00004543382,0.000001212773,0.005122339,0.9793661,0.003177804,0.0001572146,0.01094671,0.0003198004],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1591717,0.0002183051,0.8121786,0.0005055507,0.00005415698,0.0007848598,0.0001756606,0.02650317,0.0004080533],"genre_scores_gemma":[0.9697311,0.0002118084,0.02844883,0.00001554252,0.000009869625,0.0004930787,0.0009718148,0.00008573839,0.00003221355],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8105594,"threshold_uncertainty_score":0.9999754,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1036035950413029,"score_gpt":0.3214342661988822,"score_spread":0.2178306711575793,"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."}}