Research on Performance Assessment and Business-Finance Integration Based on Generative Adversarial Network Model

Against the backdrop of the rapid development of digital economy and artificial intelligence technology, performance appraisal and business-finance integration (BFI) have become key issues in improving enterprise management efficiency and scientific decision-making. Traditional performance appraisal methods often have problems such as single data dimension and subjective setting of appraisal indicators, making it difficult to effectively evaluate the synergy between finance and business. To this end, this paper proposes an intelligent performance evaluation method based on generative adversarial network (GAN) to improve the intelligence level of the appraisal system. The study uses conditional generative adversarial network (CGAN) to construct a performance evaluation framework, establishes a high-dimensional performance scoring model through the collaborative input of business data and financial data, and optimizes the rationality and accuracy of the scoring results under the adversarial training mechanism. Experiments show that compared with traditional linear regression models and decision tree models, the performance appraisal scoring error generated by GAN is reduced by 18.4%, and the performance prediction accuracy is increased by 21.7%. This method can effectively improve the scientific nature of enterprise performance appraisal, promote the in-depth development of business-finance integration, and provide theoretical support and practical guidance for enterprise intelligent management.