Here on Privacy on the Move people meet to discuss and study issues related with mobility, data mining, and privacy. Most of the material on these pages fall in the following 5 categories:
- Privacy Observatory
- Mobility Data Collection and Representation
- Mobility Data Storage
- Mobility Patterns and Pattern Mining
- Visual Analytics
WG1: Privacy Observatory
MODAP aims to coordinate research on mobility data mining, and also recognizes that technology must be disseminated, and be informed the social context in which it resides. Thus, in addition to its technical endeavors, MODAP plans to continue the privacy observatory, established in the context of GeoPKDD to bring together technologists, representatives of the national and European privacy authorities, as well as nongovernmental privacy related associations. In summary, the aim of the privacy observatory is to assist authorities as a technical consultant in the field of privacy-preserving data mining. More specifically, we believe that regulations and laws will be enacted as a response to existing and future privacy-preserving methods, including those that may be developed by the researchers within the various WGs of MODAP. The goal of the observatory is to harmonize the resulting regulations with the activities of technologists and the researchers.
The activities of the observatory will include the creation, and maintenance, of relationships with the European Commission authority and the national authorities of the countries that are partners of the consortium. Such relationships are aimed to properly implement the resulting regulations into the research methods and tools and to provide refinements of the technical regulations regarding privacy-preserving analysis methods. MODAP has a potential to interact with, and inform, organizations that recognize the need for location privacy standards. For example, one such organization is Geopriv which is an Internet Engineering Task Force (IETF) working group that examines risks associated with location-based services. The IETF has proposed several requirements for location privacy, including limited identifiability and customizable rules for controlling data flows. A second example organization is Privacy International which is a human rights group formed in 1990 as a "watchdog" on surveillance projects that are run by governments and corporations.
The observatory will have a balanced structure, where at least half of the members will be non-technical experts from law and ethics, privacy officers of member states, representatives from NGOs and similar.
At the start of the project, the observatory will consist of the WG and WP leaders within the consortium including a representative from the law and ethics community from University of Milan, a core partner of MODAP. The privacy observatory will continuously grow during the first year of the project, and at the end of the first year, we plan to reach the goal of having half of the observatory members from the ethics experts, philosophers, psychologists, human factors, legal and privacy professionals and researchers, and NGOs. At the end of the first year, the privacy observatory will be formed with the goal of the balanced distribution of technical and non-technical members. One of the main starting points to expand beyond Europe will be the Computers Freedom and Privacy organization (www.cfp.org) in US.
During the second year one of the goals of the privacy observatory will be to create a plan of sustainability beyond the time-frame of the project. One of the means of sustainability, will be through support from local governments, and sponsorship from industry.
The main roles of the observatory can be summarized as follows:
- Being the main interface of MODAP with outside world especially to non-experts.
- Technical consultation for privacy related matters.
- Producing opinion-reports on high impact hot topics such as Google street view, and its implication for privacy, new privacy regulation for geo-marketing in Germany and similar.
- Contributing to the editorial board of the MODAP news-letter.
Privacy observatory will achieve these goals through regular physical meetings, online meetings/discussions. Physical meetings will be co-located with organized events such as workshops and summer schools.
Responsible partner: Arend Ligtenberg (Wageningen University, Netherlands)
Stefan van der Spek (TU Delft, Netherlands)
Applications WG will be responsible for the new/future applications within MODAP which will motivate and direct MODAP research areas. Such applications could be adopted by high-tech SMEs and lead-to start-up companies. Possible applications of mobility data are public safety and security through better emergency response and evacuation, logistics, transportation, mixed reality games, geo-marketing, and urban planning. However these initial application topics need to be refined and elaborated with industry representatives.
WG3: Mobility Data Collection and Representation
Responsible partner: UNIL, Christine Parent
The use of powerful mobile nodes enables the creation and development of new mechanisms for storing, and sharing data and information. Nodes such as mobile phones, connected with different networking technologies (3G, WiFi, peer-to-peer mechanisms) have the capability of generating large amounts of interesting data. Such data can be transmitted to a base station, but can also be stored locally, and made available to network applications or other users through mechanisms that respect the appropriate privacy and the quality-of-service requirements. In both cases, it is important to take advantage of available local processing capabilities. For example, the cost of transmitting data can be reduced significantly if data is processed at its origin and then only intermediate, compressed data is transmitted. Depending on the nature of the data source and capabilities, different forms of data could be collected and stored. In the context of WG3, such data sources will be investigated and data collection and representation
issues will be studied which will respect privacy considerations.
WG4: Mobility Data Storage
Responsible partner: UPRC, Yannis Theodoridis
The research area of MODAP has the need for representing mobility data in databases to perform ad hoc querying, analysis, as well as data mining on them. During the last decade, there has been a lot of research ranging from data models and query languages to implementation aspects, such as efficient indexing, query processing, and optimization techniques. The realization of data models proposed in the literature as well as packaging corresponding functionality to specific technical solutions results in moving object database engines. However, privacy needs to be integrated tightly into mobility data storage and retrieval as a first step in mobility data mining. In the context of WG4, the issue of privacy preserving storage and querying of mobility data will be investigated.
WG5: Mobility Patterns and Pattern Mining
Responsible partner: NKUA, Dimitrios Gunopulos
Analysis of mobility data in the context of people's trajectories (i.e. historical collection of mobility traces of people) was initiated by the
GeoPKDD project. The aim was to define data mining models over trajectories and define algorithms to extract such models mostly in the form of mobility patterns. However, there is a plethora of different type of patterns (periodical, partial periodical) which could be discovered in an online or offline fashion from mobility data. The algorithms must be provided with:
i) scalable versions, by investigating database integration approaches and incremental learning; ii) methods for provably and measurably protecting the privacy of the data in the extracted patterns; iii) mechanisms to express constraints and queries into a data mining query language, in which the data mining tasks can be formulated. Under WG5, pattern mining research
activities from mobility data will be coordinated.
WG6: Visual Analytics
Responsible partner: FRAUNHOFER, Gennady Andrienko
Visual analytics combines automated analysis techniques with interactive visualisations in order to extend the perceptual and cognitive abilities of humans and enable them to extract useful information and derive knowledge from large and complex data and to solve complex problems. Particularly, data and problems involving mobility are inherently complex and therefore visual analytics approaches an essential component of mobility data mining. Under WG6, the problems of analysing data about movement of various discrete objects in geographical space will be investigated. Three types of movement data will be considered as a starting point which are; (1) Data describing movements of a single entity during a long time period, (2) Data about simultaneous movements of multiple unrelated entities, (3) Data about simultaneous movements of multiple related entities. The pertinent analysis
tasks significantly differ for these types of data. For each type of data, the visual analytics techniques and tools will be investigated to support mobility data mining.