@article{4386,
  author = {Lijuan Zhang},
  title = {Generalized Entropy Increase Verification and Corporate Earnings Management Based on Decision Tree Model},
  journal = {Journal of Networking Technology},
  year = {2025},
  volume = {16},
  number = {1},
  doi = {https://doi.org/10.6025/jnt/2025/16/1/1-8},
  url = {https://www.dline.info/jnt/fulltext/v16n1/jntv16n1_1.pdf},
  abstract = {With the increasing requirements of the financial market, the approval standards for listing on the stock
market have become higher, leading to many private enterprises being excluded. As a result, a method called
“backdoor listing” has become popular in the financial market for entry. To demonstrate their strength, many
companies choose to sign performance commitments before listing, with all employees committing to
performance, and then report the final performance commitment to relevant authorities for approval. However,
this approach has exposed several issues, primarily arising from the misalignment between performance
commitments and final corporate earnings. Therefore, it is essential to evaluate the performance commitments
made by employees scientifically, avoid corporate earnings management, and thereby mitigate the risk of
“performance plunge” and “performance cliff” after listing. This study collects large data using data mining
and applies cluster analysis to divide performance commitments. Finally, the relationship between employee
performance commitments and corporate earnings management impact is identified based on the decision
tree model algorithm. The results indicate that unreasonable performance commitments are one of the main
reasons for improper corporate earnings management. Using the decision tree model algorithm can make
correct decisions to reduce the risks caused by employee performance commitments. The study takes the
employee performance commitments and corporate earnings management under the state of mergers and
acquisitions as an example to elaborate on this conclusion using the algorithm.},
}