Strategy for Innovation and Development of Science and Technology

Abstract: In the context of the rapid development of information technology, huge data often brings great burden to data storage and calculation. In order to improve the efficiency and quality of audit work, this paper combines the feature selection algorithm and the big data processing technology to develop the audit data platform. By using the two-stage feature selection algorithm model which is improved on the basis of the fast correlation filter feature selection algorithm (FCBF), the characteristics are measured by the maximum mutual information coefficient, and the evaluation criteria for calculating weak correlation and redundancy is introduced. The system is oriented towards the company's audit data set, and after the characteristic selection, the innovation level of the sample is analyzed to achieve the effect of intelligent analysis.

Key words: big data, audit, two-stage feature selection, correlation, redundancy