Research And Application Of K-Anonymity In Building Data Release

With the data distribution and data mining technology is widely used in database,the disclosure of sensitive information has become a major threat to development of information technology.Sensitive information privacy protection has become an important research direction in data security and data mining applications.K-anonymity is a widely used privacy protection method in data publication.This work conduct an analysis and research on sensitive information disclosure problem in the data release in engineering building systems based on K-anonymity technology.This paper proposed K-anonymity based two-cluster generalization privacy protection model to solve the problems of existing K-anonymity in data release that it did not fully consider sensitive property.The numerical model determines the sensitivity level with the numeric difference and did clustering generalization.Meanwhile,the paper analyzed other anonymity algorithms,implemented K-anonymity based two clustering algorithm generalization on the combination of clustering and twice anonymity and provided the equations and definitions of distance and information defect.This paper use the execution time and information defection of simulation data to verity the K-anonymity based two clustering algorithm proposed in this work.In addition,the paper compared and analyzed the execution time and information defection of the proposed algorithm and P-Sensitive K-anonymity with Adult dataset.Experiment results show that the proposed two clustering algorithm generalization can protect sensitive information and ensure the potential value and statistical property of data.