{"id":"W3134250586","doi":"10.1155/2021/6685004","title":"Analysis of Travel Mode Choice in Seoul Using an Interpretable Machine Learning Approach","year":2021,"lang":"en","type":"article","venue":"Journal of Advanced Transportation","topic":"Urban Transport and Accessibility","field":"Social Sciences","cited_by":73,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Ministry of Science and ICT, South Korea; National Research Foundation of Korea; Ministry of Science, ICT and Future Planning; National Research Foundation","keywords":"Mode choice; Interpretability; Travel behavior; Multinomial logistic regression; Variable (mathematics); Mode (computer interface); Relation (database); Computer science; Econometrics; Variables; Travel survey; Logit; Gradient boosting; Logistic regression; Artificial intelligence; Machine learning; Mathematics; Random forest; Public transport; Engineering; Data mining; Transport engineering","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005339221,0.00008554271,0.0004047112,0.0002841844,0.00008546055,0.00001787872,0.0001348458,0.00006760671,0.00006367496],"category_scores_gemma":[0.00005361187,0.00008545459,0.0001990033,0.001340477,0.00005699425,0.001062451,9.408392e-7,0.0002615316,4.182318e-8],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007317199,"about_ca_system_score_gemma":0.0001872088,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002172187,"about_ca_topic_score_gemma":0.02747486,"domain_scores_codex":[0.9985113,0.0001353774,0.0006712752,0.0001590502,0.000366417,0.0001565566],"domain_scores_gemma":[0.9989983,0.00005597225,0.0004784343,0.00008480551,0.0003020697,0.00008039059],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0000567129,0.0001869817,0.5025297,0.00002044153,0.00008501071,0.000009659838,0.01270794,0.476247,0.007113779,0.00004614206,4.15689e-8,0.00099658],"study_design_scores_gemma":[0.0005691663,0.00004078247,0.9676634,0.00004949294,0.0005685888,3.871603e-7,0.007109201,0.02241163,0.001252445,0.0001894422,0.00003802068,0.000107449],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9634075,0.0003036936,0.03591936,0.00002162674,0.00007436516,0.00005469939,0.00001427504,0.000005628483,0.0001988486],"genre_scores_gemma":[0.991039,0.00006845478,0.008726487,0.00001102031,0.00003858544,6.706636e-7,0.00007297374,0.00000739239,0.00003543677],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4651337,"threshold_uncertainty_score":0.9902712,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02657846570529837,"score_gpt":0.334529398366779,"score_spread":0.3079509326614806,"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."}}