Using synonyms in SEO: bundles, holes, underpages
Virtually every SEO optimizer (commonplace SEO, website promotion, internal optimization, custom semantic core compilation, PVC windows to buy ... - as we are used to imagining, I apologize for trying to humor) encountered correlation problems of query recognition. And also with different grimaces, shortcomings in search rankings.
This short publication will be devoted to this particular problem. Where I am in the appendage will try to give a number of quite interesting examples.
We begin, perhaps, from a distance - with the fact that search engines have a large database of statistics. The power and influence of which you probably already felt. And the biggest problem was, is and will probably be “the problem of understanding a person” (which is no secret, because even we don’t really understand each other). And some banal inquiry, for example: “mittens” mislead the search engine, because it is not clear whether a person is interested in the technique of making mittens, or he wants to buy them, or he just needs a picture, or wants to know about the origin of this words, al he generally meant the oral cavity (slang). And the only correct method of solving a problem is to collect statistics and conduct analysis based on the information collected (or connect a very smart machine for reading thoughts to the equipment).
As the simplest mechanism for understanding turns, slang expressions, colloquial words (and so on), search engines begin to associate words with synonyms, as well as words-transliteration and foreign language, plus case and numerical variations (for example, "dolls", "pupae", "dolls", "Kukla", "kukly" ).
')
Attention to highlight key phrases:
However, in different search engines there are different bundles, and the PS Yandex doesn’t have, for example, dolls , for example, bundles with the phrase “dolls”. And sometimes there is, for example, as in the request "Car" :
Moreover, they are also connected in various ways, some in complete synonymous correspondence, some are not (which can, in fact, be traced by extradition), since sometimes, according to synonyms, it is also different.
There is another situation that is illogical to the pain and is a minor flaw in the Yandex search engine.
In particular, we will consider the recognition of a certain query - “German” and its relationship with the English language. After all, the difference between these phrases is obvious, right? However, Yandex looks at this situation from a slightly different angle:
Enter the query "German language" . And watch the issue:
Well, okay, with whom it does not happen, perhaps a little issue is stormed (although the request is high-frequency, what kind of storm), maybe by chance it somehow happened (although there are some accidents that 1-2 times, the system). But further more. - We did not have time to go to the second page of issue, as in the 20th place, this client meets us:
Well, okay, you might think ... Anyway, there is a phrase “German” on the main page, which is really there. But at 44 positions we are met by another beautiful site, on which there is no such phrase at all.
And all attempts to find such a coincidence inside only this site (imtp.ru) were crowned with failure (only rare single phrases and general “highlighting” of the query “English” ).
As a result, it can be concluded that the search engine “Yandex” in our case, in such a query as “German,” the word “language” considers the dominant factor (as in the type of query: “white penguin” by noun - “penguin " ), And the definition of " German "is only an insignificant secondary factor. Moreover, for such keys, the search engine even adds language variations (the same “Deutsch” for example)
It would be quite logical now to organize all this information into something sensible, which can be used in practice. Namely - the analysis of these topics will help us in building a semantic core , as well as in promoting a resource.
The webmaster or promoter is often faced with the task of choosing key phrases. However, the choice of a synonym for a key phrase can be fatal. As it often happens like this (for example, Google):
Bratz is some “puppet” brand:
"Bratz => Bratz" , but not equal to "Bratz" ;
"Bratz => Bratz" , but not equal to "Bratz" .
or how “on-line” is associated with “online” and with “on-line” , but not associated with “online” and “on-line” .;
At the same time, “online” is associated with “online” and “on line” , as well as “on-line” , but is not associated with “online” .
As a result, it is always better to choose a center like “online” , or as a way out to go along the “bass-path”, there is someone like that.
And do not forget about the complexity. After all, everyone wants to kill two birds with one stone (“Yandex”, “Google”), in this case, it is more desirable to look for averaged value among both PSs.
Similarly, there are many ways to uniqueize anchors. This is a border around a unique, near-link, and the introduction of stop words into anchors, and the imposition of temporality / incidence, and the dilution of the key on the low frequency, and the separation of keys in the links. But as an alternative method of uniquely referencing, but transferring a single utility by key, you can use phrases-synonyms, preferably those in which the output has approximately the same character, and the phrases themselves are mutually defined in both keys (often with brands: Adidas, Mercedes, Opel, and others.), often noticed that with an increase in the reference mass, at least in one of two synonymous keys, the other rises by a somewhat similar coefficient.
The post displays only my confused opinion on the above nuances in SEO or “fortune telling on the coffee grounds” (as you wish). I am pleased to take part in the discussion, as well as in the search for truth, if it is possible.
And lastly: "All SEO is reduced only to the definition of search bots" level of care "about the user." Therefore, webmasters / administrators / moderators, take care of visitors! The whole world rests only on care!
Thank you for attention.
This short publication will be devoted to this particular problem. Where I am in the appendage will try to give a number of quite interesting examples.
Synonyms and Bundles
We begin, perhaps, from a distance - with the fact that search engines have a large database of statistics. The power and influence of which you probably already felt. And the biggest problem was, is and will probably be “the problem of understanding a person” (which is no secret, because even we don’t really understand each other). And some banal inquiry, for example: “mittens” mislead the search engine, because it is not clear whether a person is interested in the technique of making mittens, or he wants to buy them, or he just needs a picture, or wants to know about the origin of this words, al he generally meant the oral cavity (slang). And the only correct method of solving a problem is to collect statistics and conduct analysis based on the information collected (or connect a very smart machine for reading thoughts to the equipment).
As the simplest mechanism for understanding turns, slang expressions, colloquial words (and so on), search engines begin to associate words with synonyms, as well as words-transliteration and foreign language, plus case and numerical variations (for example, "dolls", "pupae", "dolls", "Kukla", "kukly" ).
')
Attention to highlight key phrases:
However, in different search engines there are different bundles, and the PS Yandex doesn’t have, for example, dolls , for example, bundles with the phrase “dolls”. And sometimes there is, for example, as in the request "Car" :
Moreover, they are also connected in various ways, some in complete synonymous correspondence, some are not (which can, in fact, be traced by extradition), since sometimes, according to synonyms, it is also different.
Yandex and German
There is another situation that is illogical to the pain and is a minor flaw in the Yandex search engine.
In particular, we will consider the recognition of a certain query - “German” and its relationship with the English language. After all, the difference between these phrases is obvious, right? However, Yandex looks at this situation from a slightly different angle:
Enter the query "German language" . And watch the issue:
Well, okay, with whom it does not happen, perhaps a little issue is stormed (although the request is high-frequency, what kind of storm), maybe by chance it somehow happened (although there are some accidents that 1-2 times, the system). But further more. - We did not have time to go to the second page of issue, as in the 20th place, this client meets us:
Well, okay, you might think ... Anyway, there is a phrase “German” on the main page, which is really there. But at 44 positions we are met by another beautiful site, on which there is no such phrase at all.
And all attempts to find such a coincidence inside only this site (imtp.ru) were crowned with failure (only rare single phrases and general “highlighting” of the query “English” ).
As a result, it can be concluded that the search engine “Yandex” in our case, in such a query as “German,” the word “language” considers the dominant factor (as in the type of query: “white penguin” by noun - “penguin " ), And the definition of " German "is only an insignificant secondary factor. Moreover, for such keys, the search engine even adds language variations (the same “Deutsch” for example)
Subtleties in the selection of phrases for the semantic core
It would be quite logical now to organize all this information into something sensible, which can be used in practice. Namely - the analysis of these topics will help us in building a semantic core , as well as in promoting a resource.
Semantic core
The webmaster or promoter is often faced with the task of choosing key phrases. However, the choice of a synonym for a key phrase can be fatal. As it often happens like this (for example, Google):
Bratz is some “puppet” brand:
"Bratz => Bratz" , but not equal to "Bratz" ;
"Bratz => Bratz" , but not equal to "Bratz" .
or how “on-line” is associated with “online” and with “on-line” , but not associated with “online” and “on-line” .;
At the same time, “online” is associated with “online” and “on line” , as well as “on-line” , but is not associated with “online” .
As a result, it is always better to choose a center like “online” , or as a way out to go along the “bass-path”, there is someone like that.
And do not forget about the complexity. After all, everyone wants to kill two birds with one stone (“Yandex”, “Google”), in this case, it is more desirable to look for averaged value among both PSs.
Promotion
Similarly, there are many ways to uniqueize anchors. This is a border around a unique, near-link, and the introduction of stop words into anchors, and the imposition of temporality / incidence, and the dilution of the key on the low frequency, and the separation of keys in the links. But as an alternative method of uniquely referencing, but transferring a single utility by key, you can use phrases-synonyms, preferably those in which the output has approximately the same character, and the phrases themselves are mutually defined in both keys (often with brands: Adidas, Mercedes, Opel, and others.), often noticed that with an increase in the reference mass, at least in one of two synonymous keys, the other rises by a somewhat similar coefficient.
About the publication
The post displays only my confused opinion on the above nuances in SEO or “fortune telling on the coffee grounds” (as you wish). I am pleased to take part in the discussion, as well as in the search for truth, if it is possible.
And lastly: "All SEO is reduced only to the definition of search bots" level of care "about the user." Therefore, webmasters / administrators / moderators, take care of visitors! The whole world rests only on care!
Thank you for attention.
Source: https://habr.com/ru/post/128706/
All Articles