{"id":"W3153963714","doi":"10.1016/j.geoderma.2021.115108","title":"Enhancing the accuracy of machine learning models using the super learner technique in digital soil mapping","year":2021,"lang":"en","type":"article","venue":"Geoderma","topic":"Soil Geostatistics and Mapping","field":"Environmental Science","cited_by":120,"is_retracted":false,"has_abstract":false,"ca_institutions":"Dalhousie University","funders":"University of Maragheh; Deutsche Forschungsgemeinschaft; Alexander von Humboldt-Stiftung","keywords":"Digital soil mapping; Computer science; Covariate; Range (aeronautics); Feature (linguistics); Machine learning; Variety (cybernetics); Base (topology); Soil map; Soil science; Artificial intelligence; Data mining; Environmental science; Mathematics; Soil water; Engineering","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":[0.0003887707,0.0001135709,0.000132292,0.00002549236,0.0002125739,0.00007346122,0.0002078854,0.00004478524,0.0001907983],"category_scores_gemma":[0.0002844932,0.00007556646,0.0000515515,0.0003115733,0.0001211264,0.0002755755,0.0004315439,0.000313705,0.00001184339],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005553405,"about_ca_system_score_gemma":0.00003036452,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001242681,"about_ca_topic_score_gemma":0.0004110222,"domain_scores_codex":[0.9989402,0.00006866,0.0002685328,0.0002136016,0.0002326078,0.0002764227],"domain_scores_gemma":[0.9993444,0.0002761038,0.00009056884,0.0002433346,0.00001712508,0.00002846932],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000008037707,0.00006410822,0.0503597,0.00004012561,0.00002331884,0.00005797999,0.006513023,0.7241175,0.1996336,0.000580057,0.0000765669,0.01852597],"study_design_scores_gemma":[0.0003008536,0.00001407473,0.007756732,0.0001455473,0.00001552108,0.0001280681,0.004122785,0.9415969,0.0352318,0.00731668,0.003057449,0.00031358],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8587143,0.0001848476,0.1361665,0.0003442524,0.00004893636,0.0001774991,0.000007473166,0.00002163239,0.004334589],"genre_scores_gemma":[0.9979597,0.00003025855,0.001453824,0.0001831646,0.00002047181,0.00001370066,0.000008472033,0.00001468055,0.000315698],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2174795,"threshold_uncertainty_score":0.3081511,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02559423734149852,"score_gpt":0.2374219974620731,"score_spread":0.2118277601205746,"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."}}