Analysis of Key Paths for Big Data Technology to Improve Financial Audit Efficiency

Against the backdrop of digitalization, traditional financial audits face issues such as low efficiency, limited scope, and delayed risk identification. Big data technology offers a potential solution to these dilemmas. This study aims to systematically construct the key paths through which big data technology enhances financial audit efficiency and clarify the mechanism behind this efficiency improvement. Taking audit projects as Decision-Making Units (DMUs), the research first defines an input indicator system encompassing hardware investment, software investment, and human resource investment, as well as an output indicator system covering audit coverage rate, risk identification accuracy rate, and audit cycle reduction rate. Subsequently, the DEA-BCC model is introduced to measure the efficiency of 27 audit projects. The results show that the comprehensive technical efficiency of big data audit projects (0.892) is significantly higher than that of traditional audit projects (0.678), and this improvement is mainly achieved by enhancing pure technical efficiency (with a difference of 0.214). On the input side, problems such as hardware redundancy and mismatched human resource skills exist; on the output side, the most prominent shortcoming is audit coverage rate. The data analysis, data cleaning, and data collection modules have a significant impact on audit efficiency, with impact coefficients of 0.415, 0.352, and 0.287 respectively. The research findings can provide theoretical references and practical guidance for the digital transformation of audit institutions, helping to improve financial audit efficiency.