BIGQP’17 Workshop Description

Big Geo Data are becoming a significant part of the data production that occurs
today at a global scale. They are to a big extent crowdsourced by users who do
not follow a well-documented (« scientific ») method that ensures data quality,
either because they do not know or do not care about the issue. This kind of
data usually contain references to locations, i.e., Points of Interest (POIs),
and become accessible in general social media (e.g., Facebook, Google+) or in
specialized platforms (e.g., Open Street Maps, Yelp). Location information
could be either extracted by personal assistants (e.g., Google Now) or social
platforms (e.g., Facebook, Twitter) in terms of places visited, trajectories
pursuit or mentioned by the users, along with their social posts. Information
extraction techniques enable us to analyze a wealth of geospatial and temporal
information available in social posts such as spatial objects and the way they
are spatially, temporally and/or semantically related (e.g., north, in-between,
during, same-as). Spatial objects may refer to precise and/or imprecise
geographical objects (e.g., POIs, toponyms, and vernacular names), as well as
to implicit spatial objects identified by means of textual descriptions (for
instance, the following user post could identify a part of a certain road at
the point of publication: « traffic jam between POI ‘A’ to POI ‘B' »). The
quality of crowdsourced geo data might vary depending on the origin (machine
vs human generated), the level of detail of the extraction techniques, as well
as the obfuscation techniques used by the persons themselves or the social
media platforms to protect their privacy. Another aspect of quality is
associated with the credibility of the extracted information with respect to
one’s location or time of publication (e.g., user post mentioning an event just
after it has happened although the user’s and event’s locations are spatially
unrelated).

The quality (e.g., precision, accuracy, consistency) of geospatial information
can be improved when personal data are integrated from several data sources
(social networks, geographical authorities). On the other hand, the combination
of such personal data might reveal sensitive information regarding users’
location and might put users’ location privacy (also known as geoprivacy) at
risk. As a matter of fact, location information is inextricably linked to
personal safety. Unrestricted access to information about an individual’s
location could potentially lead to harmful encounters, for example stalking or
physical attacks. Moreover, location constrains our access to spatiotemporal
resources, like meetings, medical facilities, our homes, or even crime scenes.
Hence, it can be used to infer other personal sensitive information, such as an
individual’s political views, state of health, or personal preferences.
Understanding the different aspects of geographic/geometric/geospatial quality
involved in crowdsourced geo data and evaluate the privacy risks introduced by
enhancing their quality in personal, social and urban applications is a
challenging topic.

The BIGQP workshop aims to be a premier venue in gathering computer science and
geoscience researchers who are contributing to and are interested in both Data
Quality and Privacy of Big Geo Data. Hence, it is a unique opportunity to find
in a single place up-to-date scientific works on both subjects that have so far
only partially been addressed by different research communities such as Data
Quality Management, Distributed and Mobile Systems, and Big Data Privacy.

Topics of interest include, but are not limited to:
* Quality of online location data
* Extraction of spatial relations in Big Data
* Extraction of spatial objects from textual Big Data
* Quality metrics of Big Geo Data
* Geo entities resolution and linking
* Geo data inconsistency detection and repairing
* Geo-analytics in data quality and user privacy
* Human mobility patterns in crowdsourced Geo data
* User privacy and personal location information
* Data Quality-based Privacy models
* Privacy masking and anonymization
* Tools and Applications