Research and Implementation of a Classification Method of Industrial Big Data for Security Management

Abstract

Purpose/Significance

With the extensive adoption of cloud computing, big data, artificial intelligence, the Internet of Things, and other novel information technologies in the industrial field, the data flow in industrial companies is rapidly increasing, leading to an explosion in the total volume of data. Ensuring effective data security has become a critical concern for both national and industrial entities.

Method/Process

To tackle the challenges of classification management of industrial big data, this study proposed an Information Security Triad Assessment-Support Vector Machine (AIC-ASVM) model according to information security principles. Building on national policy requirements, FIPS 199 standards, and the ABC grading method, a comprehensive classification framework for industrial data, termed “two-layer classification, three-dimensional grading,” was developed. By integrating the concept of Data Protection Impact Assessment (DPIA) from the GDPR, the classification of large industrial data sets was accomplished using a Support Vector Machine (SVM) algorithm.

Result/Conclusion

Simulations conducted using MATLAB yielded a classification accuracy of 96.67%. Furthermore, comparisons with decision tree and random forest models demonstrated that AIC-ASVM outperforms these alternatives, significantly improving the efficiency of big data classification and the quality of security management.