Page 398 - NGTU_paper_withoutVideo

Modern Geomatics Technologies and Applications

data. Spatial data generated on the other hand is stored in different sources and access to each source generally has its own
requirements, which may be very different from the requirements of other information sources. In other words, a user is faced
with products of spatial information that are both widely distributed and structurally heterogeneous. The semantic web approach
to store and retrieve the data has the ability to solve the problems of heterogeneity and the spread of spatial information over the
Web which increase their interaction and interoperability.
The use of semantic web technology in the field of space requires standardization. The initial steps were taken by the
W3C, which provides a nominal space for displaying latitude, longitude and other information about a spatial entity using
WGS84 as the standard coordinate system. Finally, GeoSPARQL supports spatial data on the Semantic Web and defines new
terms as an acceptable W3C standard for displaying and querying.
With the rapid population expansion and by its nature, the increase in urban traffic flow, modern societies are constantly
looking for resources and control solutions during heavy traffic flows. Rainfall is one of the major factors that has disrupted
traffic flow. Ignoring this issue, in addition to the psychological burden it creates on society, imposes a number of human and
financial losses on society. For example, on January 21, 2012, heavy rains in Beijing caused more than 10 billion Yuan of
economic damage in addition to disrupting traffic in some areas of the city.
Since the beginning of this century, various activities have been carried out in the field of storage and retrieval of spatial
information and their link with other data. The authors of [3] employed the potential of data mining for different researches in a
urban traffic analysis case. They introduced data mining on linked data with the aim of answering complex environmental
2
questions, and introduced a demonstration of a traffic semantic analysis and modeling system. STAR-CITY, which integrates
sensor data into a variety of formats and volumes in the form of linked data, was designed to provide an insight and pattern of
traffic history and instant traffic.
Fonts modeled web-based search and retrieval of semantic-spatial information using existing standards such as web
3
feature service (WFS) filters[4]. Ye also provided a framework in which spatial services first define the spatial characteristics
of points of interest using semantic data, and then provide query operations and determine categories and classes[5].
Ruta have used a semantic modeling approach to store and query semantic data related to points of interest. They have
designed a system to find optimal paths applied to both semantic and spatial criteria, in addition to standardize the semantic-
spatial data collection format of R[6]. Battle and Kolas during a research to introduce GeoSPARQL, have proposed the idea that
this standard document has the potential to become a common standard between research and commercial centers[7].
Smith and partners have studied the effect of rainfall on freeway capacity and considered speed and capacity as traffic variables
and weather (rainfall volume) data. In this study one-year data has been collected from a part of the freeway while its rainfall is
classified into two modes of light and heavy. In this category, light rain is classified as 0.25 to 6.35 mm per hour and heavy rain
to that of more than 6.35 mm. This study has shown that in the light rain, the reduction of free flow rate will be up to 10% and
in heavy rain the reduction of free flow rate will be 25 to 30% [8].

The semantic web has been used in a variety of applications such as data analysis in linked data, however, information
with specific features requires more specific and practical approaches. The use of spatial and non-spatial data available in
different sources with different values and information and conversion of data into a triple graph model are among the activities
that have paid less attention in previous research. Also, data storage in graph-based databases and the use of semantic-spatial
analysis in the field of geomatics to improve urban management and in particular traffic flow improvement are among the novel
topics in this research.

This paper investigates how spatial data can be defined and interpreted in the context of the semantic web. We use
GeoSPARQL to retrieve spatial-semantic information. Investigating how rainfall affects traffic flow using the linked data and
the benefits of employing this technology are the aims of this paper. In addition, it is expected to achieve optimum and useful
information from the study of traffic critical points during rainfall in the city by spatial semantic analysis of the linked data and
its nature using databases based on some graph-based analyses.

Our journey is to achieve the stated goals of the data collection and conversion of data to Resource Description Framework
(RDF) data model using existing spatial ontologies stored in graph-based databases for the use in a semantic and spatial analysis,
to define the appropriate questions based on Spatial Data Recovery Standard - GeoSPARQL belonging to Open Geospatial

2 Semantic Traffic Analytics and Reasoning for City
3 Web Feature Service

2