{"id":"W4380986449","doi":"10.1016/j.eswa.2023.120745","title":"TDRLM: Stylometric learning for authorship verification by Topic-Debiasing","year":2023,"lang":"en","type":"article","venue":"Expert Systems with Applications","topic":"Authorship Attribution and Profiling","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":false,"ca_institutions":"Queen's University","funders":"","keywords":"Debiasing; Computer science; Artificial intelligence; Stylometry; Machine learning; Representation (politics); Natural language processing; Task (project management); Multi-task learning; Latent semantic analysis; Writing style; Information retrieval; Linguistics","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007677781,0.000146578,0.0001732042,0.0002780732,0.0006467022,0.0002611559,0.0005213956,0.0001118562,0.000003089495],"category_scores_gemma":[0.0001074548,0.0001328664,0.00004802175,0.002412892,0.00002723353,0.0002650135,0.00005324058,0.0001571351,0.0001668113],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009152037,"about_ca_system_score_gemma":0.00006310672,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003673447,"about_ca_topic_score_gemma":0.000001270301,"domain_scores_codex":[0.9984556,0.0001187594,0.0003032341,0.0004928907,0.0002724927,0.0003570473],"domain_scores_gemma":[0.9986407,0.0003369316,0.0001757419,0.0005289002,0.0001735093,0.0001441979],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00003814863,0.0001813005,0.004083182,0.0003124892,0.0001153792,0.000002765347,0.008676506,0.003550941,0.02126415,0.7981042,0.0434141,0.1202568],"study_design_scores_gemma":[0.0004793599,0.0001031467,0.0008854193,0.00005718967,0.000008668173,0.0000138434,0.0008423512,0.1702688,0.003489747,0.0007601364,0.8226488,0.0004425735],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0009880653,0.0007092693,0.9936829,0.001536601,0.0002080252,0.001157645,0.000007013069,0.0009793644,0.000731156],"genre_scores_gemma":[0.9775611,0.00003580441,0.01329715,0.0000924099,0.0002331376,0.004106754,0.0001906604,0.00002704062,0.004455946],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9803857,"threshold_uncertainty_score":0.5418134,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04554347233216235,"score_gpt":0.3100462451674291,"score_spread":0.2645027728352668,"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."}}