Keywords: GIS, WWW, digital map, attribute, natural-language processing, searching.

A primary focus of our research is the automatic creation and updating of digital map attributes by searching attribute information from Web pages, for which we have developed a system. We have also developed a system to search out spatial information using keywords based on natural language.

All world objects possess spatial information. For example, buildings and even human beings can be located by coordinate information. Thus, spatial information is closely related to our everyday life. Services providing spatial information have recently been increasing, and individuals can use these services easily via the Web or with mobile phones. An important requirement in the development of such services is the maintenance of spatial information.

Recently, maintenance of spatial information has been promoted vigorously, and a great deal of spatial information has been accumulated [1]. It has become possible, with the development of new information technology, to display spatial information in 3D form. With 3D functionality, digital maps have come to approximate the real world, but many of these maps have very little in the way of attribute information. Attributes such as names, addresses and telephone numbers are generally missing from digital maps.

Because there are so few attributes on digital maps, current services offering spatial information are not sufficient. For example, when one searches for "the Italian restaurant with atmosphere," the resulting objects displayed on the digital map should include detailed information concerning the object. Moreover, for digital maps relating to disaster prevention or environmental protection, detailed attributes are indispensable. Currently, however, attribute maintenance for digital maps requires a great amount of cost and labor.

An important requirement of attributes is that they should always be current. An effective spatial information service must provide information equal to that of the real world [2], so must continuously be updated with the latest spatial data. Such frequent updates are expensive and labor intensive, however, and this leads to delays in attribute maintenance. The development of a system to easily create these attributes is therefore very important [3].

We developed two systems for the retrieval of spatial information for digital maps. One system automatically gathers attributes associated with digital maps from the World Wide Web (WWW). Another system provides spatial information search capability based on keywords derived from natural language.

System for automatically gathering attributes automatically gathers attributes for digital maps by searching the WWW. There are six processes: 1) searching the WWW; 2) analyzing HTML; 3) extracting positional information; 4) extracting attributes; 5) acquiring coordinates; and 6) outputting attributes.

System for retrieving spatial information with natural language is used to retrieve spatial information using natural-language keywords. There are five processes: 1) entering a keyword; 2) selecting a search area; 3) creating related word and synonym references; 4) creating object references; and 5) outputting the reference results.

We have developed a system that automatically creates and updates attributes for digital maps by searching for and acquiring information from Web pages. As part of this research, we also developed a system to search for spatial information based on natural-language keywords.

The proposed system realizes the following:

• Automatic search of the WWW
• Extraction of address information and attributes by morphological analysis
• Creation of map references based on natural language

We offer the following three points concerning the social effect of this research:

• The cost of acquiring attributes for digital map creation is curtailed
• Real-time renewal of attributes is realized
• A map search engine useful to the general user is offered

From the above results, it is clear that our system is effective in extracting attributes for digital maps from Web pages.

1

Brandon Plewe, "GIS Online," Thomson Learning, 1997.

2

Roman Krzanowski, Jonathan Raper, Spatial Evolutionary Modeling, Oxford University Press, 2001.

3

T. Sagara, M. Nakayama, K. Noaki, Y. Sadahiro, "Developing WEB Yellow Page with Map Interface" Papers and Proceedings of the Geographic Information Systems Association, Vol.12, 323-326, 2003.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £95 +P&P)