Online edition of India's National Newspaper
Monday, Dec 23, 2002

About Us
Contact Us
Business
News: Front Page | National | Southern States | Other States | International | Opinion | Business | Sport | Miscellaneous |
Advts:
Classifieds | Employment | Obituary |

Business Printer Friendly Page   Send this Article to a Friend

Topic Maps

NETSPEAK THIS time takes a look at the features of the emerging technology, Topic Maps projected as a tool to retrieve information in a more efficient and meaningful way.

As mentioned several times, the problem today is not lack of information but an overload of it. Information on every conceivable subject is being generated on the Net and across various Intranets almost everyday. Locating/filtering proper and relevant information signals, weeding out irrelevant and useless information noise, from this ever-growing information base presents formidable challenges to information scientists as well as ordinary Net users.

Search engines, though helping us locate information sources, fail to give pointed/filtered information as they mechanically list out all the web pages that contain a search `string'. They have no mechanism that allows us to infer the relevance of a link before accessing the page contents. Another deficiency is that when we search for a particular topic, search engines are unable to provide us with proper sources that contain information related to the topic being searched.

We need information retrieval systems that bring out the knowledge; not just raw information; inherent in data stored at various sources. This becomes important for a knowledge-based organisation. In an organisation, different kinds of information get generated through web pages, reports, blogs, e-mails and other documents. An ordinary search result for a topic that just lists out some of the sources that contain a mention of this topic will not be sufficient. What is needed is a tool that can access the required knowledge from the multiple information sources available. Topic maps are designed to address this need.

Topic Maps: The concept of topic maps is generally explained in terms of an index — like the subject index of a book. Apart from the topics with their locations, an index gives you some information on the relationships between various topics. In some indices one might have seen entries such as `see also' and `see this topic'.

A `see also' entry against a topic means that the topic we are looking for has some relationship to the topic referred to in the `see also' entry. Likewise, `see this topic' entry suggests that our topic is the same as the topic suggested in the `see also' entry. The point to note is that an index of a book not only helps us easily navigate, it also provides valuable insights on other topics discussed in the book. So, the index in a way represents some knowledge that is external to the book's content. This also implies that the index can be used for another book with similar content; we need only to change the page numbers in which various topics occur.

This concept of index applied to the realm of digital information resources is called topic maps. Like the book index, the main constituents of a topic map are the topic, its associations with other topics and the links in which these topics occur in the various information resources such as web, company reports and manuals. These three components — Topics, Association and Occurrences — are collectively called TAO of topic maps (http://www.gca.org/papers/xmleurope2000/papers/s11-01.html).

A topic map on Theory of Relativity may contain such topics as Albert Einstein, Scientist, Physics and the like. The topic map will also show the relationships among the different topics and links to the web and other resources that contain references to these topics.

This topic map will thus provide the user with a total picture of all the topics related to Theory of Relativity and the navigation mechanism that can effortlessly take him to all the relevant resources; no need to wander here and there in the cyberspace.

Topic Map syntax standard and examples

To standardise the topic map creation process and to develop a proper topic map syntax, an organisation called TopicMaps.org (http://www.topicmaps.org) has been formed. The topic map format developed by this organisation is called XTM (XML Topic Maps — http://www.topicmaps.org/xtm/index.html). If you want to experiment with a topic map, check out http://www.ontopia.net/omnigator/models/index.jsp that hosts many examples of topic maps. This site also hosts a free topic map navigator that can be used to read topic maps from your desktop (please note that this author has not yet tested this tool).

Topic Map Resources

The TAO of Topic Maps (an excellent introduction to the subject): http://www.gca.org/papers/xmlerope 2000/ papers/s11-01.html

TopicMapmail: A mailing list where discussions on various aspects of topic maps take place. To subscribe to the list check out: http://www.infoloom.com/mailman/listinfo /topicmap mail

A list of publicly available topic maps: http://www.topicmapping.com /registry.html

A tutorial on authoring XTM topic maps: http://topicmaps.bond.edu.au/docs/6/ toc.

A Wiki site on Topic Maps: http://easytopicmaps.com

A gentle introduction to Topic Maps: http://www.topicmaps.org/xtm/1.0/ desc-intro

A Whois client

There can be many situations in which we need to know the ownership and other details of a domain name. Whois service is used to unearth the domain name details from a domain name registrar or registry's database. A Whois service scans various registry on-line databases and displays a domain name details such as its owner's postal address, technical contact, administrative contact and the domain name's server address.

There are many on-line Whois services available on the Net like ZoneEdit's Gloval Whois Gateway (http://www.zoneedit.com/whois. html) and Allwhois (http://www.allwhois.com/). If, instead of a web-based service, you need a Whois client that operates from the desktop, try out the Whois utility `Geek Tools Whois' (http://www.geektools.com/dist/gtwhois52.exe). Apart from this desktop Whois client, here you can find a link to an open source web-based Whois service (http://grove.ufl.edu/bro/cgi-bin/wp. cgi)

J. Murali

Email the author at: murali27@satyam.net.in

Printer friendly page  
Send this article to Friends by E-Mail

Business

News: Front Page | National | Southern States | Other States | International | Opinion | Business | Sport | Miscellaneous |
Advts:
Classifieds | Employment | Obituary |


The Hindu Group: Home | About Us | Copyright | Archives | Contacts | Subscription
Group Sites: The Hindu | Business Line | The Sportstar | Frontline | Home |

Copyright © 2002, The Hindu. Republication or redissemination of the contents of this screen are expressly prohibited without the written consent of The Hindu