Analyzing Enterprise Asset Structure and Management Capability Using Cloud Computing and Industrial Enterprise Financial Accounting Cost Accounting Methods

This study aims to investigate the integration of cloud computing technology within the financial cost accounting framework of industrial enterprises to enhance asset structure and management capabilities. It addresses critical challenges, including low accuracy, inefficiency, and data leakage prevalent in enterprise financial cost accounting. The study proposes a multi-user cloud financial data privacy protection and classification model based on Support Vector Machine (SVM) techniques. Initially, the model preprocesses financial data and employs a distributed two-trapdoors public-key cryptosystem (DT-PKC) for encryption, ensuring secure data uploads to the cloud platform. Subsequently, it manages multi-user keys via a trusted third-party key distribution center, thereby reducing communication costs and enhancing privacy protection. The SVM model is then extended into the encrypted domain, utilizing user public keys to safeguard sensitive information from leakage. Finally, the model trains and predicts on encrypted data using SVM, returning the results to the enterprise for decryption and informed decision-making. Performance analysis reveals that the proposed algorithm demonstrates significant improvements in financial data processing efficiency (with a runtime enhancement exceeding 10%) and accuracy (exceeding 95%) when compared to the baseline K-Nearest Neighbors (KNN) algorithm, while maintaining communication overhead below 200Kb. Consequently, this model markedly enhances the efficiency and accuracy of financial data processing, thereby strengthening enterprise asset structure and management capabilities.

本研究旨在探究云计算技术在工业企业财务成本核算框架中的集成应用，以优化企业资产结构并提升管理效能。针对当前企业财务成本核算中普遍存在的准确率偏低、效率不足以及数据泄露等关键痛点，本研究提出一种基于支持向量机（Support Vector Machine, SVM）技术的多用户云财务数据隐私保护与分类模型。该模型首先对财务数据进行预处理，并采用分布式双陷门公钥密码系统（Distributed Two-trapdoors Public-key Cryptosystem, DT-PKC）完成加密，保障数据可安全上传至云平台。随后，通过可信第三方密钥分发中心管理多用户密钥，以此降低通信成本并强化隐私保护能力。随后将支持向量机模型拓展至加密域，利用用户公钥规避敏感信息泄露风险。最后，该模型在加密数据上完成训练与预测，并将结果返回至企业进行解密，辅助企业制定科学决策。性能分析结果表明，相较于基准K近邻（K-Nearest Neighbors, KNN）算法，所提算法在财务数据处理效率（运行效率提升超10%）与准确率（准确率超95%）方面均有显著提升，且通信开销控制在200Kb以内。综上，该模型可显著提升财务数据处理的效率与准确率，进而优化企业资产结构、强化管理效能。