{"id":"W4386156233","doi":"10.32920/24034104","title":"Predicting Urban Functional Zones with Twitter Data Using the Space-Time Scan Statistics Method and the Random Forest Classifier","year":2023,"lang":"en","type":"preprint","venue":"","topic":"Human Mobility and Location-Based Analysis","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University","funders":"","keywords":"Random forest; Classifier (UML); Cluster analysis; Segmentation; Pattern recognition (psychology); Computer science; Artificial intelligence; Precision and recall; Geography; Data mining; Statistics; 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":["sts"],"consensus_categories":[],"category_scores_codex":[0.006608387,0.0002029737,0.0003610277,0.00007922699,0.001760195,0.0005289731,0.0006460833,0.0001501451,0.0005244578],"category_scores_gemma":[0.0009664849,0.0001030913,0.00007349748,0.0002831006,0.001369179,0.0001057757,0.0006092401,0.0005138052,0.00001253566],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008031426,"about_ca_system_score_gemma":0.0007418116,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.05371394,"about_ca_topic_score_gemma":0.09475576,"domain_scores_codex":[0.9969442,0.001156864,0.0003329231,0.0005923137,0.0007051313,0.0002685295],"domain_scores_gemma":[0.9941714,0.004344062,0.0002667489,0.0009096455,0.0002271785,0.00008094514],"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":[0.002211305,0.0002480598,0.3723049,0.0004881267,0.005800173,0.00001688574,0.1186272,0.2454541,0.00001673262,0.1344735,0.110654,0.00970502],"study_design_scores_gemma":[0.001238742,0.00001125825,0.007634028,0.00009170322,0.001165858,0.000001187592,0.01339138,0.9468942,6.97427e-7,0.02596041,0.003339937,0.0002705713],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01015876,0.00008866212,0.9793332,0.007201495,0.0002989807,0.0008393305,0.000222775,0.0001012787,0.001755563],"genre_scores_gemma":[0.9067996,0.000164075,0.04841154,0.0009949108,0.003732651,0.000214317,0.001509409,0.00009300563,0.03808051],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9309216,"threshold_uncertainty_score":0.9995394,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1296583238809215,"score_gpt":0.3602088202276374,"score_spread":0.2305504963467159,"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."}}