{"id":"W4366830082","doi":"10.1111/tgis.13045","title":"The challenges of integrating explainable artificial intelligence into <scp>GeoAI</scp>","year":2023,"lang":"en","type":"article","venue":"Transactions in GIS","topic":"Bayesian Modeling and Causal Inference","field":"Computer Science","cited_by":56,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; TD Bank Group","funders":"Social Sciences and Humanities Research Council of Canada","keywords":"Geospatial analysis; Scope (computer science); Computer science; Data science; Artificial intelligence; Artificial neural network; Semantics (computer science); Geography; Cartography","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0007436544,0.0001130552,0.000135719,0.0001789283,0.0002790748,0.00008445451,0.0007110023,0.00007084863,0.000004011786],"category_scores_gemma":[0.0001151753,0.00009043894,0.00006278835,0.001114939,0.000105296,0.0002658196,0.00002793222,0.0002811025,0.00005276463],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003414072,"about_ca_system_score_gemma":0.00007721694,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002590822,"about_ca_topic_score_gemma":0.0008295531,"domain_scores_codex":[0.9987813,0.00008602118,0.0003538248,0.0002721323,0.0002060639,0.0003006877],"domain_scores_gemma":[0.9987087,0.0006838451,0.00006290616,0.0004156097,0.00008340876,0.0000455586],"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":[0.000001412193,0.00004687331,0.000005741895,0.00002710416,0.000009737092,0.000005083983,0.01253337,0.008092287,0.0005430506,0.3031194,0.00002739516,0.6755886],"study_design_scores_gemma":[0.00004827917,0.0001160165,0.0001267907,0.0001462469,0.000006332984,0.000006957842,0.01539734,0.6465302,0.02115431,0.3156202,0.0007471326,0.0001002516],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.009454277,0.0004639041,0.9862892,0.001628327,0.0003218605,0.00009143445,0.000001134794,0.0001788377,0.001571043],"genre_scores_gemma":[0.9877538,0.0009645957,0.0109853,0.00001738526,0.00002232149,0.00005351574,4.173028e-7,0.000007961044,0.0001946926],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9782996,"threshold_uncertainty_score":0.3687993,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0625456185125108,"score_gpt":0.2967160792416709,"score_spread":0.2341704607291601,"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."}}