Learn About Pagerank Blog

PageRank is a patented algorithm that serves to determine which website is more important / popular. PageRank is one of the main features of the Google search engine and created by its founder, Larry Page and Sergey Brin who is a Ph.D. student Stanford University

How it works

A site will be more popular if more and more other sites that put a link that leads to a site, assuming the content those sites more useful than the content of other sites. PageRank is calculated with a scale of 0-10.

Example: A site that has a Pagerank 9 will be the first listing in the Google search list instead of sites that have a Pagerank 8 and then onwards are smaller.

Concept

Many ways to use search engines in determining the quality / ranking of a web page, from the use of META Tags, the contents of the document, the emphasis on content and many other techniques or combination of techniques that may be used. Link popularity, a technology developed to improve the shortcomings of other technologies (Meta Keywords, Meta Description), which can rigged with a special page designed for search engines or so-called doorway pages. With the algorithms 'PageRank' is, in every page will be inbound links (incoming links) and outbound links (links out) of each web page.

PageRank, has the same basic concept of link popularity, but not only take into account "the number of" inbound and outbound links. The approach used is a page will be considered important if other pages have a link to that page. A page will also become increasingly important if other pages rank (pagerank) high refers to the page.

With the approach used by PageRank, there is a recursive process in which a ranking will be determined by the ranking of a web page ranking is determined by the rankings of other web pages that have links to those pages. This process means a process that is repeated (recursive). In cyberspace, there are millions and even billions of web pages. Therefore, a web page ranking is determined from the overall link structure of web pages that exist in cyberspace. A process that is very large and complex.

Algorithm

From the approach already described in the article the concept of pagerank, Lawrence Page and Sergey Brin made pagerank algorithm as below:

Initial algorithm
PR (A) = (1-d) + d ((PR (T1) / C (T1)) + ... + (PR (Mr.) / C (Mr)))

One other published alogtima
PR (A) = (1-d) / N + d ((PR (T1) / C (T1)) + ... + (PR (Mr.) / C (Mr)))

* PR (A) is the PageRank page A
* PR (T1) is the PageRank of page T1 refers to page A
* C (T1) is the number of outgoing links (outbound links) on page T1
* D is a damping factor which can be between 0 and 1.
* N is the total number of web pages (which is indexed by google)

From the above algorithms can be seen that the pagerank is determined for each of your pages is not the whole website. Pagerank pagerank of a page is determined from the page that refers to him who is also undergoing the process of determining pagerank in the same way, so this process will be repeated until the correct results were found. But the pagerank page A is not given directly to the intended page, but previously divided by the number of links on the page T1 (outbound links), and the pagerank will be divided equally to every link on that page. So it is with every other page "Mr." which refers to the "A". After all pagerank gained from other pages that refer to the "A" are added, the value is then multiplied by the damping factor value between 0 and 1. This is done to avoid the overall value of T distributed pagerank page to page A.

Random surfer model

Random surfer model is an approach that illustrates how real that is a visitor in front of a web page. This means the chance or probability of a user clicks on a link is proportional to the number of links on the page. This approach is used so that the pagerank pagerank of incoming links (inbound links) are not directly distributed to the intended page, but divided by the number of outgoing links (outbound links), which exist on the page. It was all too thought it was fair. Because can you imagine what would happen if a page with a high ranking refers to many pages, might not be relevant PageRank technology used.

This method also has the approach that a user will not be clicking on every link on a webpage. Therefore pagerank using damping factor to reduce the value of the distributed pagerank of a page to another page. The probability of a user continues to mengkilk all links on a page is determined by the value of damping factor (d) the value between 0 and 1. Value of high damping factor means that a user will click on a page more until he moved to another page. After the user moves the page then the probability diimplemntasikan into pagerank algorithm as a constant (1-d). By removing the variable of inbound links (incoming links), then the possibility of a user to move to another page is (1-d), this will make the pagerank always be at its minimum.

In another pagerank algorithm, there is the value of N, which owns the total number of web pages, so a user has a probability of visiting a page divided by the total number of pages that exist. Sebaagai example, if a page has a pagerank of 2 and total web page 100 then in a hundred times a visit she visited that page as much as 2 times (note, this is a probability).

Related Post