{"id":"W4407114865","doi":"10.22541/au.173865299.90947187/v1","title":"Clustering Textual Features for Log Summarization in Large Software Systems","year":2025,"lang":"en","type":"preprint","venue":"","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"Natural Sciences and Engineering Research Council of Canada; Compute Canada","keywords":"Automatic summarization; Computer science; Cluster analysis; Software; Data mining; Artificial intelligence; Information retrieval; Programming language","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":[0.001262815,0.001054215,0.000773092,0.008666468,0.0007397478,0.001328606,0.0009236064,0.000852433,0.001326132],"category_scores_gemma":[0.009939871,0.0002902474,0.0007428735,0.005427059,0.0003300973,0.002127157,0.0007851121,0.0008356129,0.001510207],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005781873,"about_ca_system_score_gemma":0.0007113127,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002582626,"about_ca_topic_score_gemma":0.005679727,"domain_scores_codex":[0.9988876,0.0002693616,0.0001676007,0.0002886578,0.000297983,0.00008879706],"domain_scores_gemma":[0.9925962,0.003397295,0.001374476,0.0006725756,0.001731122,0.0002283192],"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":[0.0008203752,0.0005596129,0.02878193,0.001428729,0.0001872876,0.0004255626,0.001376189,0.02695791,0.04290113,0.001611614,0.02365955,0.8712901],"study_design_scores_gemma":[0.0001251785,0.0006064029,0.06307261,0.0002229088,0.0002897071,0.0005455599,0.002303432,0.8471895,0.04904892,0.01220594,0.02424325,0.0001465226],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3386117,0.002341385,0.6180473,0.001028945,0.0002578219,0.0009490624,0.01477187,0.02161398,0.002377928],"genre_scores_gemma":[0.5640582,0.0005891917,0.4013266,0.0001236831,0.0003024394,0.0006383304,0.03013935,0.0007132355,0.002108989],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008666468,"threshold_uncertainty_score":0.006678462,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01947680944847312,"score_gpt":0.2768511037566182,"score_spread":0.2573742943081451,"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."}}