{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001272786,0.0007398211,0.000523343,0.003256626,0.0004487802,0.001841834,0.001302535,0.0005565123,0.002632171],"category_scores_gemma":[0.008228837,0.0002865916,0.001008564,0.002150996,0.0006645191,0.003100201,0.001480746,0.001035555,0.001755428],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008271391,"about_ca_system_score_gemma":0.001305622,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005499679,"about_ca_topic_score_gemma":0.008271206,"domain_scores_codex":[0.99857,0.0004637127,0.0001542698,0.0003458858,0.0003956579,0.00007050805],"domain_scores_gemma":[0.9957937,0.002704354,0.0002804561,0.0004869636,0.0006411292,0.00009337082],"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":[0.0003763551,0.0005752245,0.007621602,0.0006158755,0.0001973152,0.0005725853,0.001143352,0.1157774,0.02134501,0.03337827,0.01543714,0.8029599],"study_design_scores_gemma":[0.00002071569,0.00005269108,0.0007256491,0.00003086052,0.00002479736,0.0001089847,0.0003168083,0.94912,0.007432749,0.03637697,0.005769492,0.00002027166],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02601902,0.00024359,0.9522918,0.0008675846,0.00006430703,0.0004267032,0.00306123,0.01351189,0.003513862],"genre_scores_gemma":[0.2961379,0.0002378785,0.6942396,0.0002505652,0.00006183828,0.0003149002,0.006348492,0.0003297587,0.002079167],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.005499679,"threshold_uncertainty_score":0.01093537,"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."}}