An Evaluation Method Based on Privacy Protection Policy for Cross-Source Data Correlation

Current correlation analysis methods often ignore the problems related to data protection during correlation evaluations between different data sources. Those methods also have some limitations on their efficiency and reliability. Therefore, this paper introduces an evaluation method based on data protection policy for cross-source data correlation. It first uses principal component analysis (PCA) to extract features from raw data, and proceeds the subsequent calculations without letting the raw data leave its original domain. It then constructs the recursive framework based on the new evaluation standards for correlation and uncorrelation to implement the evaluation of data correlation. It was proved by experiments that the method introduced by this paper could achieve the evaluation of cross-source data correlation with high efficiency and reliability while ensuring data security.