An Efficient Adaptive Graph Anonymization Framework For Incremental Data Publication

Social networks contain a wealth of valuable information, and many social networks often have huge economic benefits. In most of existing anonymous technologies in recent years, the social network is abstracted into a simple graph without time-stamped. In fact, there are a lot of incremental changes of social networks in real life and a simple graph without time-stamped can not reflect incremental social network well, so abstracting the social network into the incremental sequences becomes more realistic. In this paper, we innovatively propose an efficient and adaptive general graph anonymization framework for incremental data publication. What's more, we propose an Anonymization Process Restart Issue (APRI), and design a activated function to determine whether an anonymization process should be restarted at a certain time in order to solve the problem of high time complexity and large information loss in the incremental data publication. Our experiments on two real datasets show that the framework provides higher anonymous efficiency while keeping the high level of data utility.