This paper presents a web page fragmentation technique, which is utilized for extracting specific parts of web pages and building personalized portals using these fragments. It is based on an algorithm, which fragments a web page in discrete fragments using the page’s internal structure. A training and update procedure is used for identifying the specific fragments of a web page in different time points. Using this technique a service provider can offer Web users a system for building personalized pages based on the content of their favorite sites. This technique, besides providing a convenient way for browsing, saves time and reduces the cost of browsing in different sites.