Shulou(Shulou.com)06/02 Report--

This article mainly shows you the "example analysis of how search engines identify link cheating on the Internet". The content is simple and clear. I hope it can help you solve your doubts. Let me lead you to study and learn the article "example Analysis of how search engines identify Link cheating in the Internet".

First of all, before we talk about identifying cheating links, let's assume that when you do SEO, don't think about the ranking of search engine results in terms of possibility, feeling, or maybe. Although most SEO practitioners are struggling with search engine algorithms, it is often these possibilities, feelings, and perhaps that make a lot of SEO optimization incomprehensible. Therefore, we need to use our own hypothetical point of view, to demonstrate with a large number of data experiments, so as to get the core idea behind the ranking of results.

The simple understanding of search engines is to sort the results of user keyword requests through a manually developed program. For example, the search box in your site is actually a search engine, and the sorting methods (algorithms) of different search engines are more or less the same, but no matter which search engine it is, in a word, all have their own unique results ranking formula to rank web page results. Let's take a screenshot of a hyperlink analysis system inside Baidu, as shown below:

The above two screenshots are a screenshot of a site link of Baidu's internal hyperlink analysis system platform. From the above figure, we can learn about two core data indicators, one is the dimension of the link, and the other is the recommendation of the link.

From the screenshot above, we can disassemble the dimensions according to the assumptions mentioned above. For link cheating, search engines crack down on black links (hidden links) and link purchases, no matter which. Search engines have their own unique identification mechanism, although we do not know this mechanism for the time being, but we can demonstrate it from a hypothetical point of view.

The experimental steps and methods are as follows:

Experimental requirements: a, multiple generations of optimized random sites; B, link location consistency

The experimental hypothesis: suppose that 100 one-way links are published for 10 random sites to keep the same link position, and the speed is 3 days, and the result triggers the link cheating.

The following figure shows the comparison of comprehensive data before and after the release of 10 sites:

The link location is shown in the following figure:

According to the figure above (avoid advertising, so do not publish related sites), you can see that a set of obvious data is: domain name time, number of entries, link growth rate, link effective period, site basic link data and other dimensions determine whether the link triggers cheating. The same anchor text link location, different 10 random sites, the results are very different, and whether to trigger link cheating can be obtained according to the comprehensive data score above:

1. The ranking of links affects the ranking, and the influence of the basic data of the site triggers cheating.

2. The old domain name has a lower trigger for link cheating, while the new site has a higher degree of cheating for links in the same location.

3. Link growth rate has a great influence on link cheating.

4. Links with IP sites have a great influence on cheating.

5. The percentage of the total number of original links affects the degree of link cheating.

Then many friends must have a few questions. Through experimental comparison and data analysis, as well as two screenshots combined with Baidu hyperlink analysis system can be summarized as follows:

1. It is often said that links are released in a circular and gradual manner. Why do you release 100 one-way links in 3 days? why do you not trigger link cheating?

Strictly speaking, the gradual effect of the cycle is certainly the best. Because we are cheating on the test link, we can get two answers through the data analysis. First, the site was originally an old domain name site (there is a long time, no history of cheating), the search engine assessment of this kind of site has a special cheating mechanism adjustment. Second, as the published one-way link site (100 experimental stations) is not exactly the same time to be crawled by spiders, links need to be crawled through spider pages > link assessment (recommended) > links take effect. Different stations grab different so the effective time is also different, resulting in even if we released a large number of one-way links in 3 days, it is not easy to trigger link cheating.

2. Why did you send so many foreign chains or no rankings?

I believe that the vast majority of SEO personnel will have this question. Through Baidu hyperlink analysis data, we can find that there is no so-called weight difference between links, only the difference between effective and invalid (stop pursuing the so-called third-party weights that have no practical meaning). Content interspersed links recommendation > friendly links > cheating links (including 0 points and minus points), but because the content interspersed anchor text is crawled by spiders than the home page links, the cycle of a large number of content interspersed links to improve the ranking is longer, which is why most friends like to see whether the outer chain is included. Because of the included reasons, at least ensure that the link of the page is crawled by spiders to judge the link dimension (whether there is a degree of recommendation). Another situation is that the comprehensive data of competitors is too large, even if you send a lot of outside chain, it is not necessarily that your ranking will be very good.

3. How to avoid link cheating and improve the ranking of link effect?

From the overall four pictures above, we can draw a conclusion that the location of links needs to be diversified, such as friendly links, blog comments, navigation links, copyright links and so on. Diversified link locations will affect the calculation of links in search engines, which is why we have always emphasized the significance of extensive links. The percentage of link anchor text is very easy to trigger link cheating. Suppose a website originally has 1 million external links. If you send 10, 000 cheating links, cheating links account for 1%, which will not affect the ranking of your site. If your site chain itself is only 1000, sent 10, 000 cheating chains, the link growth data for a short period of time is 10 times, even ordinary users know that this is a link cheating. But there is a situation that even if a lot of links are posted in a short time, it will not trigger link cheating, that is, the article is reprinted. Just like I contributed to Lu Songsong's blog, with the source of my website, a large number of sites reprint Lu Songsong's articles, and some sites will also reprint my website, that is, a huge increase in internal and external chain in a short period of time. But this is undoubtedly the real recommended link for users, so it will not affect the ranking of their own site, and this mode of operation is undoubtedly the best way to avoid link cheating and improve the effect of the link. As shown in the following figure:

The above is all the contents of this article entitled "example Analysis of how search engines identify Link cheating in the Internet". Thank you for reading! I believe we all have a certain understanding, hope to share the content to help you, if you want to learn more knowledge, welcome to follow the industry information channel!

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.