hits--link algorithm based on page attributes

Intermediary transaction http://www.aliyun.com/zixun/aggregation/6858.html ">seo diagnose Taobao guest cloud host technology Hall

After reading this article, you may be able to understand the following two things:

1, a single page site, why the page is out of the chain, is pointed to a small number of links, but there is a good ranking;

2, the Site page is not out of the chain (pointing to the outside) is the best?

Do you have any interest in reading the above? Well, whether you have any interest in wood, brother will continue.

In some TV dramas, perhaps you will often see such scenes, male feet A to understand or know someone, a certain thing, often go to the local hotel hotel to waiter, or directly find the local bully to understand the situation.

Without it, just because they are like a variety of local people's Living guide, live map, they may not understand the whole story of these things, but must know the matter with the people or things involved.

The so-called investigation of the whole story, a relationship + event network.

This sort of thing on search engine algorithms, rightfully the same set up, but the role of the props all occurred in the conversion: Male pig's feet A has become a vast number of search engine users, relationship and event Network has become a link composed of various relationships, and waiter and bully became what we see today a variety of people feel puzzled site.

The hits algorithm is associated with these.

First look at Baidu Encyclopedia explanation.

Hits algorithm: An algorithm for analyzing the importance of a Web page to measure the importance of a page based on the entry of a Web page (a hyperlink to the page) and the degree (from which the page points to another page). The most intuitive meaning is that if a Web page is of high importance, the page he points to is also of great importance. An important page is indicated by another Web page, which means that the page pointing to it is of great importance. Point to another page defined as the hub value, which is pointed to as the authority value.

If you have the wood to understand, the wood relations, young continue to tell you. For any page, it must have two attributes: the directory attribute (and others like the hub) + authoritative attribute.

Directory attribute that he himself would have a chain, it will also be more like a navigation, telling the reader to see more detailed information about something that can be linked out to view; The authoritative attribute is that each page is based on a topic, and the page is bound to have authoritative values based on the subject, the size.

Learn this, and start to see how it applies to search engine algorithms.

In the previous article on search engine theory (www.seosos.cn/seo-tips/search-engine-principle.html), we introduced how the search engine handles the page and returns the result, Hits algorithm is based on this foundation. Here's a look at the detailed steps.

1. Find the Root collection

The query q is submitted to a search system based on a keyword query, and the first n pages (such as n=200) are taken from the collection of the returned results page, as the root set (root set), which is recorded as s, and S satisfies:

1.1, the number of pages in S is small;

1.2, S in the page is related to query Q pages;

1.3. The pages in S contain more authoritative (authority) pages.

2, expand related pages

With some awesome authority page, start to expand related pages.

There are only two dimensions: points to authoritative pages and authoritative pages.

To understand why this expansion of the page, you need to understand the first word: An authoritative page by multiple directory pages, indicating that the authoritative page is authoritative; a directory page points to multiple authoritative pages, indicating that this directory page is very directory.

First explain: Elder brother is not pure heart with you play tongue twister, oneself first understand.

3, calculate the authoritative value of the page and directory value

With a related subset, the above meaning is also understood, starting to compute the directory values and authoritative values for each page. The formula is simple: First, p={p1,p2,p3...pn the set of initial subsets, and then establish a matrix based on the link relationship between the pages:

If page 1 has links to page 2, the value of A12 is 1, or 0.

The authoritative value of page A is the sum of the directory values of all the directory pages that point to it;

The Directory value of page A is the sum of the authoritative values for all authoritative pages it points to.

4. Return results

Returns the results according to the page's catalog value and authoritative value.

Well, smart comrades should also be able to see that this algorithm is based on a certain theme, can be a good response to some of the characteristics of interpersonal relationships, but also a good response to some of the characteristics of the internet relationship.

It also has some of the more disgusting deficiencies, for example, time-consuming (after the return of the results of the calculation, will increase the user time to wait), there will be theme drift (regardless of the content, only to consider the link will cause is said a thing, may result into a part of the said B things) and so on.

Of course, from which we can see a fact, that is actually search engine algorithm and wood has legendary so mysterious, many are based on interpersonal relationships for the prototype calculated (the previous PR algorithm (www.seosos.cn/search-engine/pr.html) is the same), Because after all, search engine algorithmic engineer is also a person, compare, their advantage lies in how through a series of algorithms to turn this principle into machine can read reality.

This article starts with "The beauty of SEO science"

Reprint please indicate the link address: http://www.wenjuntech.com/sem/blog-1678.html