{"id":"W6901855280","doi":"10.60692/yykzm-40434","title":"Phenotyping urban built and natural environments with high-resolution satellite images and unsupervised deep learning","year":2023,"lang":"en","type":"article","venue":"Greater South Information System","topic":"Remote-Sensing Image Classification","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Deep learning; Satellite; Cluster analysis; Scale (ratio); Unsupervised learning; Natural (archaeology); Satellite imagery; Scalability; Spatial ecology","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":[0.0001628202,0.0001655639,0.0001474986,0.0002154071,0.0001558106,0.0002030497,0.00004750269,0.00006782258,0.00000103985],"category_scores_gemma":[0.00001138628,0.0001449764,0.00001438139,0.0001873552,0.00004679766,0.0007698672,0.00003008634,0.0001295591,0.000158042],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001009001,"about_ca_system_score_gemma":0.000003065772,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004650678,"about_ca_topic_score_gemma":1.816272e-7,"domain_scores_codex":[0.9991621,0.00003701996,0.0002795433,0.0001304922,0.0001783404,0.0002124855],"domain_scores_gemma":[0.9996607,0.00001063051,0.00008236045,0.0001610309,0.00002639266,0.00005892933],"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":[0.000306948,0.000004261004,0.3910617,0.005296513,0.0005066551,0.00003696503,0.3016255,0.1545299,0.01108546,0.0002634415,0.0001585871,0.1351241],"study_design_scores_gemma":[0.0006287523,0.00001561389,0.4777055,0.000152647,0.00002378647,0.00003031865,0.004079462,0.5154985,0.00135457,5.615655e-7,0.0002926627,0.0002176354],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9610124,0.00006178515,0.03753323,0.00002129486,0.0001515275,0.0002350752,0.000006648879,0.0006653388,0.000312755],"genre_scores_gemma":[0.998927,0.000008531385,0.0008388779,0.000009584282,0.00003748524,0.000008638719,0.00005910239,0.00002300693,0.00008776976],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3609686,"threshold_uncertainty_score":0.5911966,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01271755358366708,"score_gpt":0.1721694002890206,"score_spread":0.1594518467053535,"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."}}