Georgios Kellaris

Practical differential privacy via grouping and smoothing

We address one-time publishing of non-overlapping counts with e-differential privacy. These statistics are useful in a wide and important range of applications, including transactional, traffic and medical data analysis. Prior work on the topic publishes such statistics with prohibitively low utility in several practical scenarios. Towards this end, we present GS, a method that pre-processes the counts by elaborately grouping and smoothing them via averaging. This step acts as a form of preliminary perturbation that diminishes sensitivity, and enables GS to achieve e-differential privacy through low Laplace noise injection. The grouping strategy is dictated by a sampling mechanism, which minimizes the smoothing perturbation. We demonstrate the superiority of GS over its competitors, and confirm its practicality, via extensive experiments on real datasets.

Map-matched trajectory compression

The wide usage of location aware devices, such as GPS-enabled cellphones or PDAs, generates vast volumes of spatiotemporal streams of location data raising management challenges, such as efficient storage and querying. Therefore, compression techniques are inevitable also in the field of moving object databases. Related work is relatively limited and mainly driven by line simplification and data sequence compression techniques. Moreover, due to the (unavoidable) erroneous measurements from GPS devices, the problem of matching the location recordings with the underlying traffic network has recently gained the attention of the research community. So far, the proposed compression techniques have not been designed for network constrained moving objects, while on the other hand, existing map matching algorithms do not take compression aspects into consideration. In this paper, we propose solutions tackling the combined, map matched trajectory compression problem, the efficiency of which is demonstrated through an extensive experimental evaluation on offline and online trajectory data using synthetic and real trajectory datasets.

Trajectory Compression under Network Constraints

The wide usage of location aware devices, such as GPS-enabled cellphones or PDAs, generates vast volumes of spatiotemporal streams modeling objects movements, raising management challenges, such as efficient storage and querying. Therefore, compression techniques are inevitable also in the field of moving object databases. Moreover, due to erroneous measurements from GPS devices, the problem of matching the location recordings with the underlying traffic network has recently gained the attention of the research community. So far, the proposed compression techniques are not designed for network constrained moving objects, while map matching algorithms do not consider compression issues. In this paper, we propose solutions tackling the combined, map matched trajectory compression problem, the efficiency of which is demonstrated through an experimental evaluation using a real trajectory dataset.

Shortest path computation on air indexes

Shortest path computation is one of the most common queries in location-based services that involve transportation networks. Motivated by scalability challenges faced in the mobile network industry, we propose adopting the wireless broadcast model for such location-dependent applications. In this model the data are continuously transmitted on the air, while clients listen to the broadcast and process their queries locally. Although spatial problems have been considered in this environment, there exists no study on shortest path queries in road networks. We develop the first framework to compute shortest paths on the air, and demonstrate the practicality and efficiency of our techniques through experiments with real road networks and actual device specifications.

Combining Differential Privacy and PIR for Efficient Strong Location Privacy

Data privacy is a huge concern nowadays. In the context of location based services, a very important issue regards protecting the position of users issuing queries. Strong location privacy renders the user position indistinguishable from any other location. This necessitates that every query, independently of its location, should retrieve the same amount of information, determined by the query with the maximum requirements. Consequently, the processing cost and the response time are prohibitively high for datasets of realistic sizes. In this paper, we propose a novel solution that offers both strong location privacy and efficiency by adjusting the accuracy of the query results. Our framework seamlessly combines the concepts of \(\epsilon \)-differential privacy and private information retrieval (PIR), exploiting query statistics to increase efficiency without sacrificing privacy. We experimentally show that the proposed approach outperforms the current state-of-the-art by orders of magnitude, while introducing only a small bounded error.