{"id":"W7125790162","doi":"10.53106/256299802025120701003","title":"AI-Driven Tax Analytics with Transformer-Based Text Mining","year":2025,"lang":"","type":"article","venue":"International Journal of Computer Auditing","topic":"Financial Reporting and XBRL","field":"Business, Management and Accounting","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Operationalization; Audit; Database transaction; Analytics; Transaction data; Meaningful use; Narrative; Transformer; Semantic technology; Volume (thermodynamics)","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0008801181,0.0003767027,0.0006065332,0.001324956,0.0002621436,0.00121386,0.0009136081,0.0001254791,0.0001371928],"category_scores_gemma":[0.0003140908,0.0003387463,0.0004262401,0.0007324702,0.0001246546,0.001311572,0.0001130159,0.0006209501,0.00002191714],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002085036,"about_ca_system_score_gemma":0.0006750696,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005704719,"about_ca_topic_score_gemma":0.00004434668,"domain_scores_codex":[0.9962549,0.00002933871,0.001792475,0.0003282713,0.001163579,0.000431407],"domain_scores_gemma":[0.9931594,0.0002844394,0.003010747,0.0001698692,0.003335471,0.0000400708],"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":[0.001324155,0.0009273107,0.1355407,0.0007532734,0.003808861,0.003978703,0.000508542,0.2146612,0.0004535132,0.006592812,0.04678404,0.5846669],"study_design_scores_gemma":[0.006340638,0.0003423992,0.03488882,0.01351353,0.001106785,0.0003044879,0.0003923558,0.7276662,0.0006460124,0.00127332,0.212552,0.0009735183],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2423583,0.0002362251,0.7239166,0.01831864,0.006749796,0.0001683351,0.000008560199,0.00004253269,0.008200965],"genre_scores_gemma":[0.9768966,0.00001839413,0.01021671,0.006493429,0.005936916,0.000001797598,0.00001094718,0.0000355221,0.0003896292],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7345384,"threshold_uncertainty_score":0.9999065,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01013067930108178,"score_gpt":0.2546639760256312,"score_spread":0.2445332967245494,"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."}}