{"id":"W4416242390","doi":"10.1007/978-3-032-10762-6_7","title":"Trade-Off Between Interpretability and Accuracy: How Can XAI Build Trust in Track Geometry Predictive Maintenance?","year":2025,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Railway Engineering and Dynamics","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Concordia University","funders":"","keywords":"Interpretability; Support vector machine; Leverage (statistics); Track (disk drive); Artificial neural network; Bridging (networking); Precision and recall; Convolutional neural network","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004351151,0.0005254864,0.000653288,0.0008184963,0.00005846825,0.0001821382,0.0007416855,0.0004273806,0.000004038314],"category_scores_gemma":[0.000244354,0.0005139283,0.00008879032,0.0005212502,0.0004267902,0.0002180876,0.0002276749,0.001377202,6.692979e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005616636,"about_ca_system_score_gemma":0.0001425269,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001896046,"about_ca_topic_score_gemma":0.000145684,"domain_scores_codex":[0.9978313,0.00001520176,0.0003992525,0.0008478675,0.0003336903,0.0005727156],"domain_scores_gemma":[0.9984875,0.0006819074,0.00006562674,0.0005822196,0.00004121566,0.0001415247],"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":[0.000008010648,0.000009808972,0.003785952,0.0003019772,0.00003388393,0.00002932365,0.001025877,0.4384967,0.00002741823,0.0002994589,0.00002864025,0.555953],"study_design_scores_gemma":[0.0002750187,0.00007691501,0.01193662,0.0009093108,0.00001958517,0.00001420682,0.000001382904,0.978619,0.000150393,0.006775083,0.0006461421,0.0005763779],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01271878,0.0009384904,0.9829281,0.0004640086,0.001073349,0.0003997848,0.0001763256,0.000258181,0.00104299],"genre_scores_gemma":[0.9847877,0.0001820603,0.01454717,0.00009215691,0.0002190276,0.00001104004,0.00001786229,0.00004541493,0.00009753691],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.972069,"threshold_uncertainty_score":0.9997312,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.005900145420449526,"score_gpt":0.2052191868999665,"score_spread":0.199319041479517,"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."}}