{"id":"W4416828803","doi":"10.1016/j.jss.2025.112714","title":"CREST: Improving interpretability and effectiveness of troubleshooting at ericsson through criterion-Specific trouble report retrieval","year":2025,"lang":"en","type":"article","venue":"Journal of Systems and Software","topic":"Software Engineering Research","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University; Ericsson (Canada)","funders":"Telefonaktiebolaget LM Ericsson; Carleton University","keywords":"Troubleshooting; Interpretability; Relevance (law); Software; Volume (thermodynamics); Data retrieval; Human–computer information retrieval","routes":{"ca_aff":true,"ca_fund":true,"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":[0.003097396,0.002293007,0.001507723,0.003553943,0.0005057122,0.002665495,0.002614225,0.00171527,0.002185293],"category_scores_gemma":[0.01617927,0.0004058753,0.001362565,0.001448841,0.0007260107,0.002932405,0.002334557,0.001586905,0.002572367],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001151306,"about_ca_system_score_gemma":0.002633389,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01832409,"about_ca_topic_score_gemma":0.01475354,"domain_scores_codex":[0.996807,0.000743243,0.0002643635,0.0006907972,0.00122915,0.0002654007],"domain_scores_gemma":[0.9932175,0.00256764,0.0006274185,0.001244727,0.002092897,0.0002498044],"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":[0.001519746,0.0005999263,0.01364688,0.000678344,0.0002869142,0.001080213,0.001277854,0.2170868,0.04075024,0.003843812,0.02089505,0.6983343],"study_design_scores_gemma":[0.00007689407,0.0003694458,0.002828008,0.00004103583,0.00007730487,0.0002717788,0.0003484889,0.9671741,0.0210869,0.002822829,0.00482046,0.00008275895],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2191583,0.001464068,0.7326926,0.0006765943,0.00016276,0.0005356647,0.001501997,0.03904708,0.004760976],"genre_scores_gemma":[0.6487065,0.0005350302,0.3373719,0.0003403987,0.00008826243,0.0002067012,0.006287988,0.001064525,0.005398782],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01832409,"threshold_uncertainty_score":0.03643483,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01474976466855962,"score_gpt":0.2766366461209478,"score_spread":0.2618868814523882,"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."}}