Personalized trajectory anonymization through sensitive location points hiding

The omnipresent use of GPS enabled smart devices left numerous mobility traces in an unprecedented scale. The analysis and publication of these ordered spatio-temporal points is essential for the inventions and discovery of new mobility management applications. But the publication of trajectory details is certainly a privacy threat to the object/or individuals. So a privacy preserved anonymization approach for the publication is needed. It is observed that, instead of anonymizing the whole trajectory, the sensitive location samples are to be protected. This work proposes a new model, which extracts the delicate halting points using Haversine distance measure and protects these points from adversary attack by a personalized generalization technique. This model also mitigates the exposure of whole trajectory by safeguarding the sensitive location points in a diversified area zone. From the evaluation point of view, this model uses the real-world data set and the results show that the model is more effective against malevolent attacks and is having less information loss than the existing well-known anonymity approaches.