Geography Reference
In-Depth Information
Chapter 10
Creating Spatiotemporal Semantic Maps
from Web Text Documents
Wei Wang and Kathleen Stewart
10.1
Introduction
With the continued development of the Internet, text information is widely available
in the form of Web articles, news reports, blogs, Twitter feeds, and other formats.
Because of the massive amount of information provided by unstructured texts, the
need to acquire relevant information automatically from text sources is necessary
for web users. Geographic information is commonly referenced in these Web
documents. In addition, many documents describe dynamic occurrences (e.g., the
track of a storm) of natural phenomena such as severe storms as well as human
activities (e.g., the movement of relief supplies in the wake of a natural disaster).
This chapter discusses methods for combining natural language processing tech-
niques and ontologies to improve the extraction of semantics of hazard-related
events sourced from web texts. Developing methods to systematically represent
the movement of dynamic entities such as severe storms and their associated
spatiotemporal behaviors is important for next generation information system design
where details extracted automatically from text can be presented to users for their
own domain applications.
Information retrieval (IR) provides valuable opportunities for users to obtain
information automatically from Internet search engines or digital library applica-
tions. For example, when a user types “2012 London Olympics” into a search engine
(e.g., Google), the query terms are used for accessing relevant web documents
that contain this text string using IR algorithms, and a set of documents related
to London Olympics will be retrieved. While the field of traditional IR has
contributed solutions for helping users to automatically find information based
;
Search WWH ::




Custom Search