The explosive day-to-day growth of information available on the web has necessity the web users to make use of some techniques to locate desired information from web resources.
Web contains noisy data, redundant information and which mirrored web pages in and abundance.
The effective way of identifying required patterns is a major issue the necessity to discover data from web sources and needs to be address.
In this paper we propose an efficient method to address some of the problems during web content extraction.
In the proposed method we extract required patterns by removing noise that is present in the web document.
Proposed method shows better performance when compared with existing methods.
In future we plan to extend our work to construct DOM tree (Graphical representation) after extraction of useful patterns.