{"id":"W4310496924","doi":"10.1016/j.artmed.2022.102471","title":"Why did AI get this one wrong? — Tree-based explanations of machine learning model predictions","year":2022,"lang":"en","type":"article","venue":"Artificial Intelligence in Medicine","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":52,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"European Commission","keywords":"Computer science; Machine learning; Artificial intelligence; Boosting (machine learning); Decision tree; Ensemble learning; Fidelity; Gradient boosting; Random forest","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.009719279,0.0006851691,0.0007300162,0.001352191,0.0007432863,0.002791493,0.00198158,0.002493857,0.004174839],"category_scores_gemma":[0.07906243,0.0004690576,0.0009302305,0.000959538,0.002174797,0.005351392,0.002206113,0.004104683,0.0008263438],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001166413,"about_ca_system_score_gemma":0.001207364,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001297624,"about_ca_topic_score_gemma":0.001730931,"domain_scores_codex":[0.9946724,0.003517268,0.0002051022,0.0005510763,0.000888119,0.0001660702],"domain_scores_gemma":[0.9567192,0.03189374,0.003351112,0.004857404,0.002666801,0.0005116959],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0006428587,0.000157432,0.0248691,0.0006376674,0.0003643618,0.0008670406,0.005018233,0.1643354,0.00475014,0.5497095,0.01828044,0.230368],"study_design_scores_gemma":[0.00006080468,0.0000931022,0.002557071,0.0002086517,0.00004752679,0.0003001334,0.0004362118,0.6284562,0.002939624,0.354667,0.01016232,0.0000714143],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03863977,0.0007039066,0.947149,0.007920483,0.0002498629,0.00006971075,0.000368505,0.001223307,0.003675461],"genre_scores_gemma":[0.60475,0.0005982181,0.3898367,0.001343281,0.0002761343,0.0001591768,0.0006140419,0.0004341316,0.00198838],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.009719279,"threshold_uncertainty_score":0.05140108,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0677420181985633,"score_gpt":0.3151730733071335,"score_spread":0.2474310551085702,"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."}}