{"id":"W4415757169","doi":"10.2139/ssrn.5688858","title":"Transit Pulse: Advancing Public Transit Social Media Analysis with Large Language Models","year":2025,"lang":"","type":"preprint","venue":"SSRN Electronic Journal","topic":"Human Mobility and Location-Based Analysis","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"","keywords":"Public transport; Social media; Analytics; Pipeline (software); Topic model; Transit (satellite); Big data; Action (physics)","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":[0.004322851,0.00160324,0.001621378,0.002551737,0.001122927,0.003445891,0.001663968,0.001509938,0.00937824],"category_scores_gemma":[0.02982218,0.0008015029,0.002445586,0.002199437,0.0007352813,0.005940332,0.003203028,0.003892766,0.004758787],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001029556,"about_ca_system_score_gemma":0.002192794,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01719644,"about_ca_topic_score_gemma":0.0219375,"domain_scores_codex":[0.997568,0.001422351,0.00009885688,0.0003807224,0.0004213131,0.0001087942],"domain_scores_gemma":[0.9784008,0.01807054,0.0004477474,0.001680109,0.001026571,0.0003741768],"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":[0.001127312,0.001838878,0.0198076,0.001170054,0.002871839,0.0009695375,0.001597994,0.2276364,0.004773874,0.1225824,0.1510952,0.4645289],"study_design_scores_gemma":[0.00005456587,0.00003690594,0.0006384423,0.00003080914,0.00009094254,0.00003142458,0.0000990465,0.9474476,0.0003564589,0.0450848,0.006109747,0.00001932773],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02865961,0.0008930457,0.9433344,0.003642862,0.0005398502,0.0001506541,0.008299386,0.01188697,0.002593301],"genre_scores_gemma":[0.4489411,0.001809203,0.5055619,0.001423675,0.001833664,0.001032349,0.02638067,0.003869962,0.009147477],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01719644,"threshold_uncertainty_score":0.03419268,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01181682854351028,"score_gpt":0.2773954569491393,"score_spread":0.265578628405629,"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."}}