Random forest model in tax risk identification of real estate enterprise income tax.

ABSTRACT: The text describes improvements made to the random forest model to enhance its distinctiveness in addressing tax risks within the real estate industry, thereby tackling issues related to tax losses. Firstly, the paper introduces the potential application of the random forest model in identifying tax risks. Subsequently, the experimental analysis focuses on the selection of indicators for tax risk. Finally, the paper develops and utilizes actual taxpayer data to test a risk identification model, confirming its effectiveness. The experimental results indicate that the model's output report includes basic taxpayer information, a summary of tax compliance risks, value-added tax refund situations, directions of suspicious items, and detailed information on common indicators. This paper comprehe

SUBMITTER: Xu C 

PROVIDER: S-EPMC10965060 | biostudies-literature | 2024

REPOSITORIES: biostudies-literature