{"id":"W4301396626","doi":"10.1002/geot.202200047","title":"Practical recommendations for machine learning in underground rock engineering – On algorithm development, data balancing, and input variable selection","year":2022,"lang":"en","type":"article","venue":"Geomechanics and Tunnelling","topic":"Dam Engineering and Safety","field":"Engineering","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"York University","funders":"","keywords":"Toolbox; Machine learning; Computer science; Algorithm; Process (computing); Artificial intelligence; Resource (disambiguation); Data mining","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":[0.000575826,0.0001413722,0.00014558,0.0001607337,0.0002552417,0.00005432351,0.00007715469,0.00004658621,0.00001574432],"category_scores_gemma":[0.00005761433,0.0001740778,0.00000979708,0.0002171904,9.605349e-7,0.0001437754,0.0001243586,0.0004389664,8.388175e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001035986,"about_ca_system_score_gemma":0.00002657948,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001646594,"about_ca_topic_score_gemma":0.000009836102,"domain_scores_codex":[0.9991626,0.00001497871,0.0002074484,0.0002611753,0.00009202697,0.0002617595],"domain_scores_gemma":[0.9996112,0.0001812582,0.00002660002,0.0001096625,0.00001273953,0.00005856848],"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":[0.000006487146,0.00002174804,0.00002667558,0.00007033612,0.00003628967,0.000001512565,0.000128824,0.9857562,0.0003179874,0.004260933,0.0001748463,0.009198215],"study_design_scores_gemma":[0.0003064727,0.00004726252,0.00001120229,0.00002650653,0.000009375546,0.00002769874,0.00007339386,0.9031233,0.00005978484,0.0003280312,0.09580968,0.0001772577],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001691548,0.000197949,0.9972032,0.000125684,0.0003852911,0.0001421117,0.0000432093,0.0001845875,0.00002644673],"genre_scores_gemma":[0.6529798,0.0005496831,0.3449616,0.00006750142,0.0001673058,0.0001295284,0.000855629,0.0001149339,0.0001740594],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6522416,"threshold_uncertainty_score":0.7098686,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02771238770067212,"score_gpt":0.2427019547295757,"score_spread":0.2149895670289036,"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."}}