Types of web mining

The world wide web, www or web is becoming a complex universe. Naturally, deriving something valuable out of it is targeted use of web mining. The application of data mining techniques in to the web and by that process deriving a higher sense of the mess is called web mining. When we say web here, it may be the content; contents like text, images, audio, video etc., that numerous web sites are holding. Or it may the structure of the web, like hyper links, tags etc., or it may be the usage statistics or logs.

Web mining consists of these three areas:

• Content mining
• Structure mining
• Usage mining.

Traditional data mining methods and techniques are used to do the web mining.

Typical Google or Yahoo or Microsoft Bing search that we do, and the resultant links listing page we get is an example of content mining. The process of extracting useful information from the web content happens here. The search is done by search engine which includes a spider. The search program delivers the result with the co-ordinated effect of search engine. The search can be for text or image or multimedia. Search Engine Optimization – SEO – techniques are used to increase the delivery of web page to the search engine queries.

Web structure mining is done at the hyper link level. A relevant example can be Google’s Page rank. HITS and Pagerank are applied web structure mining uses. We know from this web structure mining process the authority of a particular web page and its hyper link structure. This is also used for categorisation of different websites.

Web usage mining process involves the log time of pages. The world’s largest portal like yahoo, msn etc., needs a lot of insights from the behaviour of their users’ web visits. Without this usage reports, it will be difficult to structure their monetization efforts. Usage mining has direct impact on businesses

Let us see a real time example of web structure mining. Alexa is one of the leading web information sites. Compete is a competitor for Alexa. Alexa Traffic Rank is presented in web site for the top ranking sites. If we take an example of a site, ehow.com, and search the same in Alexa we get details of its rank for worldwide and for countries and complete traffic statistics. We can note that between 2009 and 2010 the traffic to this site has increased fivefold and 1 percent of the global internet users are visiting this site. Compete is a similar web site to Alexa.

The growth of e-commerce, online shopping is adding fuel to the web mining technologies. In fact among all the potential use of web mining in future, the growing online shopping activities, e-services industry, e-commerce are important domains. Particularly web usage mining where customer log in and manoeuvrability details are analyzed and resulting patterns found.

About the author:

• Malick Md
• PMP, Technology Expert And Writer
• Read complete bio
• Become an ExpertsTown Writer