L–Diversity-Based Semantic Anonymaztion for Data Publishing

Nowadays, publishing data publically is an\udimportant for many purposes especially for scientific\udresearch. Publishing this data in its raw form make it\udvulnerable to privacy attacks. Therefore, there is a need to\udapply suitable privacy preserving techniques on the\udpublished data. K-anonymity and L-diversity are well\udknown techniques for data privacy preserving. These\udtechniques cannot face the similarity attack on the data\udprivacy because they did consider the semantic relation\udbetween the sensitive attributes of the data. In this paper,\uda semantic anonymization approach is proposed. This\udapproach is based on the Domain based of semantic rules\udand the data owner rules to overcome the similarity\udattacks. The approach is enhanced privacy preserving\udtechniques to prevent similarity attack and have been\udimplemented and tested. The results shows that the\udsemantic anonymization increase the privacy level and\uddecreases the data utility