{"id":"W3196207475","doi":"10.1289/isee.2021.p-098","title":"Spatiotemporal characterization of urban activity and environment with imagery and deep learning","year":2021,"lang":"en","type":"article","venue":"ISEE Conference Abstracts","topic":"Urban Transport and Accessibility","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; University of British Columbia","funders":"","keywords":"Bespoke; Convolutional neural network; Built environment; Geography; Correlation; Business; Computer science; Cartography; Artificial intelligence; Ecology; Advertising; Biology","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":[],"consensus_categories":[],"category_scores_codex":[0.0001890301,0.00007568733,0.0001330761,0.0000193763,0.0001493757,0.00006966439,0.00004250614,0.00005733478,0.0001414282],"category_scores_gemma":[0.00003255438,0.00007153527,0.0000130732,0.00005366902,0.000280721,0.0004355788,0.00001551629,0.0001094128,0.000001452551],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000136963,"about_ca_system_score_gemma":0.000113485,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004527402,"about_ca_topic_score_gemma":0.0007883252,"domain_scores_codex":[0.9992951,0.00006201557,0.000126294,0.000210574,0.0001757475,0.0001302516],"domain_scores_gemma":[0.9995996,0.00004171066,0.000138623,0.0000815335,0.00005774173,0.00008076675],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00005236292,0.0001261023,0.830521,0.0000680477,0.00001871894,0.00002674218,0.007730976,0.00002791132,0.1155806,0.0002441026,9.239178e-7,0.04560253],"study_design_scores_gemma":[0.0001381329,0.00002314102,0.9880975,0.00002563748,0.00001376517,3.876654e-7,0.0004416528,0.00001465912,0.01058257,0.00008448611,0.0004913189,0.00008668811],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.997511,0.00006313989,0.0004518467,0.0002919699,0.00002053859,0.0000882003,0.000004506951,0.00001578362,0.001553015],"genre_scores_gemma":[0.9992786,0.0002444634,0.0001210852,0.0000117601,0.00003026522,0.000002858102,0.00002012178,0.000004459845,0.0002863813],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1575766,"threshold_uncertainty_score":0.2917123,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01764277347488803,"score_gpt":0.2411545986265231,"score_spread":0.223511825151635,"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."}}