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

This article shares with you the content of a sample analysis of SEO optimization techniques. The editor thinks it is very practical, so share it with you as a reference and follow the editor to have a look.

How to set SEO keywords is of course important, but how boring it would be if SEO optimization technology only mechanically said what to do and what not to do, and the steps involved were very rigid, such as setting focus keywords on title 1, key words repeated three times in the first paragraph, and so on.

In the past two decades, search engine optimization is based on keywords because computers cannot effectively analyze natural language input to understand the purpose of search users, and then rely on search users to enter simple phrases as guidance, which is called keywords. Keyword research Keyword Research refers to the use of tools, such as Google Adwords Planner, to analyze the search volume and competition of individual keywords. Because the traditional search engine optimization takes keywords as the object, a basket of keywords will be assigned to different pages within the site as targeted keywords according to their relevance to the content.

Basic concepts of SEO keyword

Knowing more about the basic concepts and basic theories of keyword setting will make SEO optimization more interesting. For example, the basic theory of the occurrence density of SEO keywords is TF-IDF. The search engine wants to determine the relevance Relevance of the SEO keyword and content from the keyword density that appears on the page. Using plain language to exclude mathematics is not difficult to understand:

The frequency with which keywords appear on a web page is TF (Term Frequency). If the keyword K appears on the page four times, the TF value should seem to be 4.

The problem is that if page An is twice as long as page B, you cannot assert that the keyword K is more relevant to page An if the keyword K appears twice as often on page An as page B. So TF is modified to the number of occurrences / length. For example, there are 4 times when the file has a chapter of 100 words:

TF value = 4 / 100 = 0.04

The problem is that, for example, in the description of an article introducing "online business", some words are like "customer", and the TF of "online" is obviously higher than that of "online business". However, it is obviously against common sense to say that the article is more relevant to "online" or "business" than to "online business". When the article is about "online business", "cloud computing", "online payment", "online teaching" and so on, "online" is likely to be mentioned. The conclusion is that the word "online" is more common in general articles. Only using TF to calculate the correlation will appear the above deviation.

The solution is to give the more popular keywords the lower the weight Weight. For example, in 10000 articles, the keyword "online" appears 100 times, and the weight becomes reciprocal (so called IDF or Inverse Document Frequency). For example, if the keyword "online promotion" appears once in 10000 articles, the weight becomes 10000 Inverse Document Frequency 1 = 10000. In order to ensure that the weight value does not change hastily, it is calculated on the basis of log base 10.

"online" IDF value = log (10000 / 100) = 2

"online promotion" IDF value = log (10000 / 1) = 5

Multiplying TF by IDF is the score of the correlation. Of course, TF-IDF is more complex in today's search engines, but the principle is the same. Today, search engines have evolved to be very complex. In addition to the senior engineers who are engaged in SEO research within Google, people who are engaged in SEO optimization outside can only try to understand the essence according to some symbols such as patent applications and different explanations of Google personnel. The algorithm used by Google search engine to calculate keyword correlation is based on TF-IDF, but it has evolved to be more advanced.

Long tail keyword

The long-tailed keyword Long tail keyword refers to the product, service or information you are looking for by adding more words before or after or at the same time based on the general keyword. The long-tailed keyword has a high IDF value. For example: "large size clothes for weddings", "anti-sensitive cosmetics" and so on. The long-tailed keyword is characterized by a small number of searches, but the websites that provide products, services or information related to the keyword are also small, so the search ranking of the long-tailed keyword is less competitive, making it easier to get high search rankings.

Focus keyword

The focus keyword Focus Keyword means that the web content takes this keyword as the ranking goal, and the IDF value of the focus keyword is low. So the focus keyword is added to the important elements in the page to inform the search engine of the relevance of the web page content and the focus keyword. The ranking optimization recommendations of SEO optimization tools such as Yoast SEO and SEOpressor need to set focus keywords first. In general, web page content should only aim at one or two focus keywords, otherwise it may lead to a lack of sense and reduce readability.

Single-page correlation or multi-page correlation

Some websites accept Guest Posting posts from tourists, and it is not as important that a single page has a high correlation with a keyword as it is with several pages showing high relevance. Using ITF, you can detect the correlation between the entire site (or, more accurately, the pages indexed by search engines) and a keyword. For example, the ITF of "Korean Fashion" in a general website is 0.0003, that is, only 3 pages out of 10, 000 pages are mentioned, while that of a certain website is 0.05, that is, five pages out of 100 pages are mentioned, which means that the website is very likely to take "Korean Fashion" as its theme. That is, the website has a high authority on "Korean Fashion", that is, the An in the high-quality content E.A.T described by Google Rater Guideline, which stands for Authority. Search engines prefer highly authoritative content.

Direct or indirect correlation

Keyword has a synonym Synonyms. For example, teachers, teachers, teachers represent very similar or even exactly the same concepts or things. When the correlation between the web page or website and keyword An is high, and keyword B is synonymous with keyword A, the correlation between the web page or website and keyword B is also high. However, because it is indirectly related, the ranking score of the search input keyword B is slightly lower than that of the directly related keyword A.

Whether it should be directly related or indirectly related depends on the geographical location and cultural differences. Some synonyms are caused by cultural and geographical differences. For example, "moggie" is a British term for mixed domestic cats, that is, we usually call them "cat". As the popularity of mobile phones can more accurately determine the geographical location of users, British users can be judged as cats by typing "moggie" (search engines are likely to judge that "cat" is directly related to "cat"), but Chinese users may not type "moggie" (search engines are likely to judge that "cat" is indirectly related).

The conversion of complex and simple keywords can also be regarded as indirectly related, so you will see that simplified ranking results sometimes appear when entering traditional keywords. But it is also because of the indirect correlation, so the ranking score is slightly disadvantageous.

The relevance of potential word meaning

With the increasing popularity of mobile phones, the use of voice search Voice Search will increase significantly. When users slowly find that the original search engine's understanding of natural voice input is not bad, the search user behavior will also change, instead of phrase keyword input to more natural sentences. The emergence of RankBrain is to deal with more complex meaning of words, through the use of artificial intelligence AI to understand the potential meaning and to build a relationship between the user's search input and implicit keywords. For example, the user inquires what is wrong with PDF and translates it into "PDF security" through the analysis of the underlying meaning of RankBrain.

LSI and Proximity

The meaning represented by a single word is sometimes very vague. For example, "Apple" stands for fruit and widely refers to a listed company in the United States. So the exact correlation judgment needs to look at the above, for example, if the web page is about fruits and apples, it is likely to find all the keywords about food in the text, such as "health" and "taste". If it's about Apple, you can find keywords such as "company", popular products such as "iPhone" and so on. For example, in the information on "psychological counselling", most of them will mention the word "emotion". Lincoln in English can be the name of a person, a car, or the president of the United States. Latent semantic index Latent Semantic Indexing (LSI) means that search engines judge the relevance of keywords according to other relevant words mentioned above. These related keywords should appear not too far away from the main keyword, and the farther apart means the weaker the relationship between the two.

SEO keyword density

From the above theoretical introduction, you can see that the occurrence of more keywords is not of great help to the search engine to judge the relevance, on the contrary, some long-tailed keywords may be more conducive to IDF weight scoring.

Because keyword density has been abused. Nowadays, the density of keywords appearing on the web page has become unimportant. Instead, we pay more attention to the frequency of LSI dark correlation. The next step after understanding the basic concepts of the SEO keyword is how to choose the keyword.

Keyword position

Keyword location Keyword Position refers to the effect of keyword location in different elements on search rankings, especially when elements appear in places of limited length, such as headings. Generally speaking, the more the position appears, the better, but sometimes putting the keyword in front of the keyword may not make sense and will reduce the search user's desire to click.

Thank you for reading! This is the end of this article on "sample Analysis of SEO Optimization Technology". I hope the above content can be of some help to you, so that you can learn more knowledge. if you think the article is good, you can share it out for more people to see!

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.