{"id":"W4413451213","doi":"10.1016/j.jss.2025.112594","title":"Hybrid approach for multilevel multi-class requirement classification: Impact of stop-word removal and data augmentation","year":2025,"lang":"en","type":"article","venue":"Journal of Systems and Software","topic":"Topic Modeling","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Saskatchewan","funders":"Natural Sciences and Engineering Research Council of Canada; University of Saskatchewan","keywords":"Class (philosophy); Word (group theory); Computer science; Artificial intelligence; Data mining; Natural language processing; Engineering; Mathematics","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":[0.002592547,0.001169541,0.00181182,0.002532689,0.0009244701,0.002056726,0.002619165,0.00183188,0.00391219],"category_scores_gemma":[0.00646647,0.0004064415,0.002024239,0.002424628,0.0003438898,0.002354686,0.001793597,0.002668653,0.003065397],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005672551,"about_ca_system_score_gemma":0.002385994,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009497127,"about_ca_topic_score_gemma":0.01893274,"domain_scores_codex":[0.9976307,0.000576381,0.0002270448,0.0006558945,0.0006200555,0.0002898582],"domain_scores_gemma":[0.9942391,0.00288988,0.0002137774,0.0008284961,0.001588308,0.0002404321],"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":[0.001462245,0.001738605,0.01038895,0.0005282054,0.0005098797,0.0002008878,0.0005471581,0.02875733,0.03472622,0.002086482,0.01208831,0.9069658],"study_design_scores_gemma":[0.00007082018,0.0002229367,0.005211799,0.00004194744,0.0002213005,0.0001400087,0.0004046745,0.9703718,0.01472739,0.002994203,0.005534693,0.00005829249],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1370241,0.001432495,0.845178,0.000918382,0.0004069218,0.0003991188,0.002138427,0.009457503,0.003045044],"genre_scores_gemma":[0.4429042,0.0003218639,0.5442976,0.0004244677,0.0002032443,0.0004261532,0.006080738,0.0005832149,0.004758577],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009497127,"threshold_uncertainty_score":0.01888371,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.133419085261096,"score_gpt":0.3621934879905279,"score_spread":0.2287744027294318,"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."}}