{"id":"W4400614524","doi":"10.1016/j.compscitech.2024.110759","title":"Explainable artificial intelligence prediction of defect characterization in composite materials","year":2024,"lang":"en","type":"article","venue":"Composites Science and Technology","topic":"Thermography and Photoacoustic Techniques","field":"Engineering","cited_by":32,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of British Columbia, Okanagan Campus; University of British Columbia","funders":"Texas A and M University","keywords":"Thermography; Finite element method; Materials science; Transfer of learning; Benchmark (surveying); Heat transfer; Characterization (materials science); Computer science; Inverse problem; Void (composites); Inverse; Nondestructive testing; Artificial intelligence; Machine learning; Infrared; Structural engineering; Composite material; Optics; Mechanics; Mathematics; 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":[0.0002465772,0.00007219252,0.0001089711,0.0008120969,0.00004674289,0.00004861006,0.0001409584,0.00007639689,0.000007529964],"category_scores_gemma":[0.00001127396,0.00006994985,0.0000109008,0.00150503,0.0003578897,0.0001914049,0.00003925033,0.00008193706,8.58591e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002598798,"about_ca_system_score_gemma":0.00001539685,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000005585362,"about_ca_topic_score_gemma":0.000001341381,"domain_scores_codex":[0.9994278,0.000006693499,0.0001814432,0.0001553506,0.00008226074,0.0001464421],"domain_scores_gemma":[0.9998002,0.00002242331,0.00001450273,0.0001050821,0.00004105152,0.00001671044],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00000288301,0.000008788241,0.0004874532,0.00008368104,0.000003255159,0.00000345345,0.0001103722,0.00003755822,0.9684749,0.01335974,0.00000135762,0.01742662],"study_design_scores_gemma":[0.00001011535,0.00005774024,0.001761194,0.000141263,0.000005702583,0.00001736957,0.00003407945,0.0181859,0.9717267,0.007950035,0.00005208285,0.00005781024],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9727985,0.000386839,0.02570947,0.00003975295,0.0001947233,0.0001340167,0.00009284608,0.0005496195,0.00009419935],"genre_scores_gemma":[0.9993467,0.0001195756,0.0004804554,0.000003080945,0.00001087908,0.00002165576,0.0000100969,0.000006368462,0.000001186908],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02654817,"threshold_uncertainty_score":0.2852472,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.00991524094260036,"score_gpt":0.2169606090643569,"score_spread":0.2070453681217565,"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."}}