GLOSSARY - Mobility Data - page 363

identifier: In a database, an attribute (or a combination of attributes) whose value uniquely
identifies objects. For example, each value of a Social Security Number uniquely identifies
the person it is associated to. This is to be contrasted with quasi-identifier .
knowledge discovery: The process of extracting useful and nontrivial knowledge from
data. It includes three main steps: data pre-processing , data mining ,and data post-pro-
cessing . When applied to trajectory data it is often referred to as mobility knowledge
discovery.
lifting: In spatio-temporal databases, a technique used to derive operations for data types
varying on space or time from the operations on the corresponding base data types by
allowing each argument type to be space or time dependent. For example, a function to
compute the distance between two fixed points is lifted to obtain the computation of the
distance between two points, fixed or moving.
location-based service: An information service accessible through a mobile device that
makes use of the geographical position of the mobile user to determine the most appropriate
answer to a user's query.
location k -anonymity: A privacy paradigm for the protection of the mobile user's identity.
Auseris k -anonymous with respect to position if his/her position is indistinguishable from
the position of at least k − 1 other users.
location prediction: A predictive model specific to moving objects that is able to forecast
the future locations that the object will visit. These models are usually built from the history
of past behaviors.
location privacy: An information privacy concern that addresses the protection of personal
location information.
map matching: For objects moving within a network, the process of combining the
recorded location of the object with the digital map of the network to obtain the real
position of the object within the network.
measure: In data warehouses , a metric that quantifies facts in a cube . For analysis, mea-
sures are aggregated along the dimensions of the cube. For example, a measure that states
the price at which a product is sold in a given branch of a retail store can be aggregated along
a branch dimension to obtain the average retail price of the product among all branches.
movement track: The sequence of raw data representing the movement of an object for
the whole duration of the movement.
online analytical processing (OLAP): Interactive analysis of data contained in a data
warehouse . It comprises a set of operations such as drill down, roll up, slice, and dice.
ontology: In computer science, a formal representation of a set of concepts within a domain
and the relationships between these concepts. The formal representation is equipped with
an inference mechanism to perform logical inferences on the ontology.
origin-destination matrix (OD-matrix): A representation of flows in the form of a matrix
where the rows and columns correspond to different locations and the cells contain aggre-
gated values from the attributes of individual trajectories.
pattern: A representation that characterizes a set of data in a summarized way. In data
mining , a pattern is a model that represents a summary of the analyzed data set with respect
to some criteria. See also trajectory behavior .
point of interest (POI): A specific location that is of interest in a particular context.
Examples include monuments, hotels, and restaurants. Notice that point of interest is
a generic term which does not necessarily mean that the specific location has a point
geometry; it can be a line or a region. It is also referred to as place of interest.