{"id":"W4408238568","doi":"10.7763/ijcte.2025.v17.1366","title":"Enhancing Tag Recommendation Precision on Stack Overflow Data Warehouse: An Integrated Approach Combining Numeric Attributes, Feature Extraction Techniques, and Multiple Machine Learning Algorithms","year":2025,"lang":"en","type":"article","venue":"International Journal of Computer Theory and Engineering","topic":"Data Mining Algorithms and Applications","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Windsor","funders":"","keywords":"Computer science; Stack (abstract data type); Data mining; Data warehouse; Algorithm; Machine learning; Feature (linguistics); Artificial intelligence; Operating system","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":[0.005410874,0.001210247,0.001844366,0.008777403,0.000992809,0.003162091,0.001345823,0.0009629513,0.0006513462],"category_scores_gemma":[0.01426118,0.0004344116,0.001245964,0.007776226,0.0003195772,0.003840688,0.001290489,0.001041988,0.000936683],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001092508,"about_ca_system_score_gemma":0.001728544,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01421641,"about_ca_topic_score_gemma":0.022106,"domain_scores_codex":[0.9973429,0.0005653818,0.0003481367,0.0005572942,0.0009949548,0.0001913601],"domain_scores_gemma":[0.9889025,0.004761142,0.0009198466,0.001953681,0.003173675,0.0002891213],"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":[0.000883304,0.0005994996,0.05836117,0.000338458,0.0004157754,0.0002336731,0.000598903,0.03240456,0.02151429,0.00159952,0.006912666,0.8761382],"study_design_scores_gemma":[0.00006832807,0.0003846005,0.02380399,0.00009891987,0.0002944683,0.0002767585,0.0006993276,0.923463,0.03821195,0.005131761,0.007411209,0.0001555826],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3460659,0.001702554,0.6276982,0.0006378582,0.0001570665,0.0003216765,0.0031111,0.01679528,0.003510382],"genre_scores_gemma":[0.4830093,0.0003610058,0.5119959,0.0001483058,0.00006830619,0.000110853,0.002953804,0.0001635667,0.001188911],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01421641,"threshold_uncertainty_score":0.02861583,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01649627597894825,"score_gpt":0.2781303409130938,"score_spread":0.2616340649341455,"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."}}