Аннотация: Article on actual topic of the site advance
THE CONCEPTION OF CONSTRUCTION OF OPTIMAL WEB-SITES ON BASE OF PAGE RANK
(Article on actual topic of the site advance)
Actually, the World Wide Web is some new planet, like the planet of people. This new planet is populated with human documents, web-sites.
The Earth helps us to understand ourselves, for the Earth resists us. The man is known in his clearing of obstacles. In the same way the WWW helps us to understand ourselves. The Web-sites, like animated bodies, take shape, develop, dye, enter the coalitions, compete against each other.
In other terms, the web-site is a virtual projection of a man or a company to the Net.
Our main task is as follows: to make the web-site maximum viable, to understand the site-promoting correctly and to embody its principles into the real life of the WWW.
It is important the site to have a good rating in the Search System, i. e. to take one of the top places in the Search replies to the Users' requests.
Rating on the site-promoting, the Search Systems of the Internet resources are an important source of, as what you might call, the traffic of any web-site.
Hereinafter the narration shall be oriented to the Google Search system for the reason of its leading position among the Search Systems.
For the correct methodical development and understanding of the problems of construction of the optimal web-site, and also for understanding of the anatomy of the large-scale hyper-text Google Search system, one should put a question: what is the index of a document (of a site page).
This is the most important thing, to achieve which the Search Robot Googlebot works.
One might summarily say, that the index of a document is its annotation. From the methodical point of view, this will not be correct. A site is the object for processing for the Googlebot.
The web-site for the Googlebot is a guide multigraph, on the top of which the pages are placed. The Search system Robot investigates the site and constructs its model.
In the model the main content of the site is depicted.
Thus, the index is the model of the site examined and, formally, it is set by the structure which concludes the following: texts (parts of the texts), titles, heads, anchor texts, parts of the body texts, of the hypertexts of the site pages.
In other terms, all elements of the hypertexts are presented, and what is very important, the search units are formed in tabloid form; the above search units one should call the search concepts: K (title), K (head), K(anchor), K (body).
Certainly, the Page Rank takes a special place in this structure.
The Page Rank permits to estimate the site topology. The Page rank sets estimating values for the pages, included to the site.
Proceeding from these estimations, the Search Robot Googlebot pays attention to every page included in the site in a different measure. Moreover, the part of the pages can be left ignored by reason of their law estimation. But even, if there are numerous pages, connected by the intensive mutual references (intensive interlinking), the PR technology shall assume every page the same estimation (1) and the Googlebot shall reserve the right to process only the part of these pages.
Conclusion: it is necessary to fulfill the internal linking, having set such structure of links between the site pages, that the estimating values of the most important pages to predominate explicitly the estimating values of the rest (additional) pages.
To speaking in images, the guide multigraph of the site is associated with a skeleton, on which flesh, i.e. the text and other data, is fixed.
Mathematically, the computational scheme of the PR is the recursive procedure, the particularity of which is that the outlet values, received at the certain stage of iteration, shall become the input ones of the next iteration.
It is important that no matter what outlet values are (on can even set zeroes) the recursive procedure, after execution of the final number of the iteration cycles, shall calculate the limiting values. The characteristic feature of the limiting estimating values is that these values are not changed in the further iterations, and the iteration process will not be completed at this level.
Obviously, when the search Robot indexes the site it uses the values available as the starting ones.
The recursive procedure permits to judge about the changes on the site, which could have taken place from the moment of the last indexation.
The situation, when the search system, not having set the recursive procedure to the limiting values, shall reserve the intermediate results and come back for the further calculation to the other sИance of indexation, is possible.
It is very important for the Googlebot, especially on analyzing of external links. Such approach permits to check to see that the new external links are stable, but not timely and bluff, organized by a commercial deal.
The equation of the recursive computating scheme.
PR (A) is the Page Rank of page A,
PR (T1) is the Page Rank of a site (page) pointing to Page A
С1 (T1) is the number of the individual links off Page T1 to Page A,
Сr (T1) is the general number of links off that Page T1,