Car Traffic Monitoring - Mobility Data - page 220

collective mobility knowledge, implemented in activity-based models capable
of supporting the decisions of mobility and transportation managers.
10.6 Bibliographic Notes
The analytical scenarios presented in this chapter are linked to the techniques
and methods presented in the previous chapters. In this section we will provide
a general list of references to the scientific literature for the reader.
The analytical process for mobility data is based on a specific instance of
the knowledge discovery process ( Giannotti and Pedreschi , 2008 ), where ana-
lytical methods and algorithms are composed by means of an SQL-based lan-
guage ( Trasarti et al. , 2011 ), introduced in Chapter 7 , and integrated in the
analytical framework of M-Atlas ( Giannotti et al. , 2011 ).
The estimation of travel demand by means of the four-step model is presented
in Ruiter and Ben-Akiva ( 1978 ). The basic concept of this approach is the
definition of an origin-destination matrix where rows and columns represent
zones of origin and destination respectively and each cell estimates the flows
between the two corresponding zones. This model has been extensively used
in mobility data management to select, aggregate, and analyze specific traffic
flows. Chapter 8 presents an overview of different methods to visualize and
interact with OD matrices.
The mining algorithms, trajectory pattern, clustering, and WhereNext were
introduced in Chapter 6 . In this chapter, we adopted a clustering process based
on the progressive clustering approach ( Rinzivillo et al. , 2008 ), where the
clustering analysis is organized in a stepwise process.
The extraction of the borders of human mobility by means of network analyt-
ics methods was originally presented in Brockmann et al. ( 2006 ), where mobility
flows are measured by observing the movements of banknotes. Successive works
adopted a similar approach using telephone usage data ( Ratti et al. , 2010 )and
GPSdata( Rinzivillo et al. , 2012 ). The identification of groups of nodes within
a network is performed with a community discovery method. An extensive pre-
sentation of the available community discovery methods is given in Coscia et al.
( 2011 ). Chapter 15 presents several techniques to analyze mobility by exploiting
network analytics methods.