{"id":"W4322620236","doi":"10.1108/f-08-2022-0119","title":"Using image analysis to quantify defects and prioritize repairs in built-up roofs","year":2023,"lang":"en","type":"article","venue":"Facilities","topic":"Infrastructure Maintenance and Monitoring","field":"Engineering","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Convolutional neural network; Asset management; Roof; Facility management; Computer science; Originality; Asset (computer security); Artificial intelligence; Engineering; Civil 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004383885,0.0006629099,0.0002989159,0.002133345,0.0001655825,0.0009891032,0.0004330288,0.0004712676,0.001320419],"category_scores_gemma":[0.0008446673,0.0002362932,0.0004486835,0.0005563805,0.0003483526,0.0006848528,0.0004395793,0.0002696326,0.0004178267],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006298048,"about_ca_system_score_gemma":0.0004177146,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005462219,"about_ca_topic_score_gemma":0.008701622,"domain_scores_codex":[0.9996756,0.00003390013,0.00001803137,0.00009859364,0.0001203412,0.00005352796],"domain_scores_gemma":[0.9995661,0.00006938386,0.0001300529,0.00004874888,0.0001594097,0.00002633986],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0004150042,0.0002525873,0.1222072,0.0005795053,0.0002244111,0.0004345827,0.0002920527,0.1462303,0.1893633,0.001736719,0.002910019,0.5353544],"study_design_scores_gemma":[0.00001008085,0.0002153368,0.1300431,0.0001332797,0.0001292553,0.0003566247,0.0003851785,0.7591516,0.103957,0.00147349,0.004100202,0.00004493759],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6613582,0.001061404,0.3273592,0.0002414231,0.00008529862,0.000172781,0.001137045,0.001582879,0.007001804],"genre_scores_gemma":[0.9280185,0.0003520196,0.06903317,0.00005377139,0.0000173037,0.00003078204,0.0005461939,0.00004838117,0.00189998],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.005462219,"threshold_uncertainty_score":0.01086086,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02757931862359225,"score_gpt":0.2814650142651397,"score_spread":0.2538856956415474,"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."}}