We all know that if you want your website to rank high, you must do SEO, and you must know how search engines work. In fact, it is not difficult to understand the principle. Search engines on the market are about the same. Take Google as an example.
Google uses two important features to get accurate query results: first, Google uses the link structure of a webpage to calculate the ranking of each page, which is called PageRank; second, Google uses link information to further improve search results.
The basic idea of PageRank is: If a web page is also pointed to by other web pages multiple times, it means that the web page is more important or of higher quality. In addition to considering the number of page links, Google also needs to refer to the level of the linked page itself and the number of forward links between that page and other pages. Of course, links to “important” pages will have a higher weight. PageRank’s simplified calculation formula:
PR (A) = (1-d) + d (PR (T1) / C (T1) + … + PR (Tn) / C (Tn))
• PR (A): PageRank value of page A;
• PR (Ti): PageRank value of page Ti linked to page A;
• C (Ti): the number of outbound links on the web page Ti;
• d: damping coefficient, 0 <d <1, usually set to 0.85
PageRank can be obtained by combining the vector matrix with the weights of the vector matrix. (In order to analyze PageRank in depth, I will write another article at my convenience.)
Random surfing model:
PageRank can be understood as a user’s behavior model. Let us assume that a random website visitor “random surfer” gives a random webpage, he will continue to click on the links in the webpage until he is tired and browse the new random webpage from scratch. PageRank can be understood as the probability of a web page being randomly accessed. The damping coefficient d is the probability that a random visitor does not follow a link to a web page and performs a random surfing from the beginning. For some webpages, the damping coefficient may be artificially changed, which can prevent some cheating websites from misleading Google to obtain a higher PageRank.
You can also consciously understand PageRank like this: Pages with a higher PageRank are pages that point to many pages, or pages that point to important pages. Google assumes that if a page is referenced by many other different pages, it means that the page is also worth a look. Also, if the page points to a site like Yahoo, it’s usually worth a look.
Link description text
Google has special treatment for connection description text. Most search engines associate link text with the page they are on, and Google also associates link text with the document to which it points. The reason for this is that link descriptions often provide a more accurate description of the web page to which they point.
In addition to PageRank and link descriptions, Google also uses some other features: First, Google records the location information (hits) of all keywords, and uses the keyword’s correlation analysis entirely in search. Second, Google records some visual information, such as font size. Large and bold fonts have a higher weight than other fonts on the web.
In addition, Google believes that textual information not directly provided to visitors may be misused and misled by search engines. So Google pays less attention to the text of metadata.