{"id":"W4411613687","doi":"10.1016/j.eswa.2025.128731","title":"Contrastive and self-supervised learning for open-set damage classification in structural health monitoring with incomplete and imbalanced vibration data","year":2025,"lang":"en","type":"article","venue":"Expert Systems with Applications","topic":"Structural Health Monitoring Techniques","field":"Engineering","cited_by":8,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"Natural Science Foundation of Jiangsu Province; KU Leuven","keywords":"Computer science; Structural health monitoring; Artificial intelligence; Set (abstract data type); Machine learning; Open set; Data set; Vibration; Supervised learning; Pattern recognition (psychology); Structural engineering; Mathematics; Artificial neural network; 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":[0.0002006159,0.0001711169,0.0002740209,0.000113441,0.0002893811,0.000156689,0.0002728233,0.00006192606,2.060091e-7],"category_scores_gemma":[0.00001090985,0.0001462391,0.000004581874,0.0002684233,0.00003571103,0.0003781173,0.00008133932,0.0001563211,1.428463e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001967034,"about_ca_system_score_gemma":0.00007181143,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005349282,"about_ca_topic_score_gemma":0.00007907231,"domain_scores_codex":[0.9989431,0.00004604908,0.0003103457,0.0003880291,0.00009056231,0.000221925],"domain_scores_gemma":[0.9992129,0.0001481231,0.00009421072,0.0004014843,0.00006775869,0.00007555843],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001123557,0.0001531174,0.6299744,0.01065263,0.00073247,0.000004590003,0.02645512,0.01653347,0.03071609,0.04733639,0.001807007,0.2345112],"study_design_scores_gemma":[0.00173577,0.0001636669,0.3718666,0.0007216441,0.00001334499,0.00001200299,0.002896764,0.6187984,0.0004081225,0.0001314206,0.002913626,0.0003387279],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.594051,0.003983558,0.3874274,0.0008320568,0.0002640414,0.01178532,0.000142721,0.001300655,0.0002131788],"genre_scores_gemma":[0.9560744,0.0001707719,0.04040219,0.00001629693,0.00009856486,0.003029422,0.0001726321,0.00002798323,0.000007742457],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6022649,"threshold_uncertainty_score":0.5963457,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04166158333456722,"score_gpt":0.3453157598843263,"score_spread":0.3036541765497591,"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."}}