How do search engines work? The basic principles of the search engines

Search engines.

The replenishment of information resources of the Internet is proceeding at a high pace, and it is becoming increasingly difficult to find the necessary information. Various printed guides become obsolete even before publication. The only reliable way to search for information is to use various search engines that constantly monitor changes in information on the network. During the existence of the Internet, various attempts have been made to organize search tools. Many of these attempts were unsuccessful, while others led to the creation of convenient means of finding information. The most successful projects have appeared in the last five years. We will look at information retrieval on the World Wide Web using some of the most common search engines. In total, there are hundreds of different search engines in the world, and the choice of a particular system depends only on your personal preferences. Note that search engines are often referred to as search engines or search engines.

Understanding Keyword Search

Select the most popular web pages among these pages to indicate higher quality. Sort pages with information closest to the specified keywords. . Come on, how to find a mortgage. Know that the broker is most able to find the best loan offers.

Internet search follows a generalized process that leads us through the operation of search engines. The meaning of keywords in itself is no longer one parameter: it is not as simple as before, otherwise too many searches are lost in accuracy, since the content on the network sometimes looks like different things. For a search engine, choosing no longer means finding words that match the pages of a website.

Many search engines allow you to search for information not only in Web pages, but also in newsgroups and file repositories. Thus, as a result of the search, you can find a message in a newsgroup or some file. Therefore, more often used instead of the term page is a more general term - a document. By document is meant a web page, message or file containing various information.

The popularity of a real estate broker site greatly affects its rating. Indeed, search engines realized that a quality search by users should lead to the most visited sites and pages, often synonymous with the quality of the content. This is a virtuous circle: the better the site, the more people will return or be attracted by its content, and thus, the more content is visited, the better it will be.

However, these same search engines provide valuable tips to increase the likelihood that they will be listed first in search of potential customers. These free tips are effective for correctly linking to your site, but there are more sophisticated methods to put your brokerage firm in the forefront.

The search process is quite simple: by asking keywords specific to the information you are looking for, we will find the document we need. Regardless of which search engine you use, an exemplary information retrieval algorithm is as follows.

Ø You go to the start page of the search engine or to any page on which there are fields for entering a query and a button for starting a search. In the latter case, after clicking the button, you will automatically go to the search engine page. There are no fundamental differences between the search engine itself and the page with fields for entering a query that is passed to the search engine. The main thing is to go to the page where you can enter your request.

Make sure you have index credit content

Search engines, although complex, are limited in their search for the vast Internet, and they do not see what we see. As you can see above, images, animations, videos, and other non-text content are not viewed by the search engine. In even more detail: Accompany the images with hidden descriptive text that describes the photo, but will not be human-readable, only a search engine. Write a transcript for your audio and video content that will contain a written record of what only a person can understand.

Ø If the system allows sequential refinements of the search area, then you go to the sections that define the boundaries of the information search (for example, search only in the field of science or search among European Internet sites). In such systems, the query in the main section searches among all Internet nodes. If, for example, you go to the "News" section, then a search by request will be performed only among sites dedicated to news. In the event that you go to the weather section, you will only search among the weather news. Thus, you first determine the boundaries in which to conduct a search, and only after that a request is sent to search for information. If the search system you selected does not allow you to refine the query, this action is skipped.

Duplicate content and canonicalization

Duplicate content is a common problem on many sites, which dramatically reduces search engine rankings. This can happen if you offer a regular page with a page optimized for printing. The use of keywords is fundamental in the search process. Highlight the keywords of your profession and your services in the text, but do not overdo it, because search engines recognize an excessive concentration of keywords, which can significantly reduce your ranking.

Ø A search query is entered in accordance with the rules adopted in the selected search system. In the simplest case, this is one or more words, for example, “Jokes”, but complex queries with logical operations “AND”, “OR”, “NO” are also possible.

Ø If the search system allows, then some query parameters are refined using additional fields, lists, flags and radio buttons, which can be located on the page next to the query field. In most cases, parameter refinement is not required. Refining parameters and compiling complex queries is somewhat similar to programming and is more popular with programmers than with a wide range of people who are not familiar with programming.

Shipping link structure

Be clear and understandable to your audience. Never use keywords in a text anchor that goes to another site, which avoids the cannibalization of keywords. Create a network on your sites between pages to be integrated into your text to create a “network” of related content. However, if you ask visitors to your site to fill out a secure form, most likely the search engine will never see this content. Finally, too many links on related pages impede navigation in search engines.

Ø A button is pressed to start the search, and after a pause, you will automatically be taken to a page with a list of documents satisfying your request. The length of the pause depends on the complexity of the request, the speed of the search system and the quality of your connection with this system.

Ø If there are many links and they do not fit on one page, then after reading the first page with the search results, you can go to the next, and so on. Thus, you look at all found links to documents.

If you want root bots to protect your site, be warned that you also do not allow the search engine to access it. If you do not have sufficient knowledge of these structures, stay as far as possible. The search option for your site does not allow the search engine to view your content.

The use of "rich fragments"

The names of your site

  Use hyphens to separate words rather than characters, or, rather. . They must meet the requirements of search engines that prefer short, precise headlines. They must contain the keywords you want to click on search engines. They must also contain your brand in order to highlight it in the search engine results.

Ø After finding the desired link, you click on it and go to the page you were looking for. If you want to look at other pages found, you can go back to the page with links to the documents found and go to another link.

In many search engines, the query field remains on the page while viewing the results, and you can quickly refine the query and organize a new search. In most systems, the list includes, in addition to links, several lines of each of the documents found, so that it is easier to select the right one.

Accessibility for all users, even for search engines

It is usually said that cheater referrers strive to achieve high rankings in search engines by any means possible, while Honest optimizers strive to encode their web pages in accordance with search engines. After you read for several weeks and carefully corrected my personal site to fit most of me, there is a huge similarity between high availability and the most effective “honest” abstracting methods. If you think about it further, this coincidence makes sense.

In some systems, link lists are sorted so that at the top of the list you will get links to documents that most satisfy your request. For example, if a document contains the query words many times and several words included in the request are located nearby in the document, then such a document more fully satisfies the request. Since most systems perform intelligent search on the Internet, the degree to which the search engine matches the query will not always match your ideas.

Go through several control points

The purpose of accessibility is to make website content accessible to as many people as possible, including those who are under technical, physical or other restrictions. These are the types of problems that are supposed to be addressed first. Search engines are also deaf when it comes to audio files. Providing textual descriptions to these files here also allows search engines to better interpret and classify content that they cannot “hear”.

It should be noted that when performing a query, the search system does not search for a document directly on the Internet. She turns to her database, which collects data on information on the Internet in a compact form. It is possible that the real state of the Internet is somewhat different from what is stored in the database. Search engines are constantly updating and updating their databases to minimize this mismatch.

Text links are very important for search engines because link text often indicates, concisely, the content of its landing page. In fact, many reviewers believe that link call text is a major factor in modern search algorithms. If the site uses a clickable image rather than a text menu as the main navigation method, an additional text menu elsewhere on the page will provide search engines with additional information about the content of each landing page.

Major search engines support country-specific indexes. Specifying the language of the document helps search engines determine in which index to place them. Similarly, search engines "browsers" do not read scripts; For this reason, the usability of the web page should not be reduced due to the lack of script support.

How does a search engine work?

Almost every search engine (let's say - the classic search engine) consists of three main components:

Despite the exotic nature of this name, a web spider is just a special program that runs on a computer (computers) connected to the Internet, and its main task is to wool all this Internet (more precisely, Internet pages) in all possible directions . After all, the pages, as we said, consist of hyperlinks, and so the spider rushes like crazy about the "web" sites hyperlinks and downloads the pages thus obtained for the second component - the index base.

Otherwise, search engines may have problems with indexing the page and not even talk about its ranking. But if a website uses “the purest and simplest language for its content,” it probably uses keywords that will be most obvious to potential search engines. Researchers tend to use short sentences sentences using familiar language.

Therefore, in order to get the maximum traffic from search engines, it is preferable that the site contains words that the site’s audience will use during their searches. For example, milestones 2 and 5 relate to the availability of dynamic content.

An indexer is a handler for pages downloaded by a web spider. He extracts all the words from there and puts them in a search database (index database). At the same time, the indexer records exactly where a particular word was found, and this information is then used in the search.

The search algorithm is the main know-how of any search engine. The effectiveness of the result depends on the algorithm - that is, how quickly and accurately the user will find what interests him.

Design and Content Tips

In fact, the conversion of dynamic content to “readable” content by search engines is one of the most burdensome tasks that the referee encounters when working on an e-commerce site or on a database-driven site. Strive to define a clear hierarchy between pages and create effective text links. Each page must be accessible from at least one static text link. If some features prevent your entire site from appearing in a text browser, robots may encounter similar crawling difficulties.

  • Create a site map that links to the most important sections of your site.
  • To check the operation of your site, use a text browser.
Perhaps they are afraid to scare site administrators with technical jargon?

Attention! Thus, when a user enters his query, the search engine searches for an answer in its index database and displays the results in accordance with its search algorithm.

For a good search engine to work, all three components are important. Moreover, each of them is actually very complicated, and its work obeys a huge number of all kinds of tricky rules, which are also constantly being corrected.

Of course, for most web designers, the goal of web accessibility is to make sites accessible to everyone, regardless of platform or disability. The good news is that a web designer who follows good Internet access practices already practices reliable, “honest” links.

The expanded tool has been available to all English-speaking users of the social network since July, which is a limitation inherent in the complexity of the search system for understanding the human language, since “Search by graph” answers a specific question and not simple keywords. For France, the site has only the anticipation of "Scheduled Search."

The search engine should have a nimble spider, a powerful index base, and an effective search algorithm - only then will you get really useful results.

You can talk for a long time about how to form search queries, but I prefer to reduce the main recommendations to the following simple postulates.

Socialized Search Engine

In short, socialized research. For example: "People Living in Paris." Once upon a time, you have access to thousands of products: these are “links”, in other words, you have the opportunity to find them. When you walk on the shelves, you see that some products have greater visibility than others: this is because they are “positioned”.

And if people in marketing work well, this positioning will generate more sales. Your site may be present in the search engine, but your personal interest will be that it is positioned, that is, it is located in a strategic place, so that it is easier to find when the user is searching. The first thing to understand is that there is a real difference between links and positioning.

1. Write correctly the words of the search query

You will laugh, but a lot of people can’t find sites on the topics they are interested in and search engines are busted up in every way, while the search word is simply entered with an error, and the search engine wipes the hard drives to the holes, trying to find the “urgent tailoring shop” ”, While in the atelier no one sews but sews a coat.

The goal of positioning will be to optimize your pages according to relevance criteria so that your site leaves the party and is the best possible within the search engine. Of course, before working on positioning, you need to work with the site. This is the first step.

To be present in the search engine, you just need to fill out a form with a search engine or even create a link on the Internet that leads to your site. On the other hand, to be well represented and stand out, it is more tiring! If you are not internationally oriented, let's say it makes you worry less. Otherwise, you will need to start by searching which search engines are used in the country you are targeting.

2. Use synonyms

If the search for the desired results did not bring, try reformulating the query using synonyms. For example, instead of “chopper” write “cool motorcycle”, instead of “car” - “car”, instead of “term paper” - “abstract”.

3. Specify the request

The more accurately a search query is built, the more likely it is that the resource you need will be in the first lines of the search result. Therefore, if you are looking for the already mentioned technology of cross-stitching on milk cans, simply and stupidly write this in the search bar. If the result does not satisfy you, then already begin to vary the line.

Each search engine has its own list of relevance criteria that may affect your positioning. Search engines logically want to highlight the highest quality websites, and this requires criteria.


In addition, a social strategy can be implemented.

  • If your site is regularly updated.
  • If your site has good traffic.
  • If your site is recommended by peers.
  • If your site is quickly displayed.
  • If your site is compatible with tablets and smartphones.
Search engines want to promote quality sites that are relevant to the user. It helps if you have to deal with big competitors, but you can do without it.

4. Use keywords

If the search result did not satisfy you, include as many qualifying words as possible in the search query. If you are interested in the work of Herbert von Karayan, then in the search bar, besides the name of the great conductor, be sure to enter the word "creativity". If you are looking for information on a Honda car, then write "Honda car", and not just "car".

5. Do not write the request in uppercase

It is advisable to write all the queries in lower case, because the search is usually case-sensitive, and the string “FIVE MILANIAN DEPARTMENTS” will be found only if it is typed in capital letters somewhere on the site. However, if you are looking for some proper names - for example, the Black Coffee group, and not a black coffee food product - then write them in capital letters (namely, with capital letters, and not all with capital letters).

6. In difficult cases, use the query language

Almost all search engines support the so-called query language, which allows you to set powerful combinations of various search criteria.

Query Language

Suppose you want to ask the search engine the following way: “Find me all the pages where the word“ herring ”appears, but in any case the word“ potato ”is missing, no more than two words from the“ herring ”is the word“ vodka ” so that the site is in the ru zone and the document itself refers to www.exler.ru ”. Good inquiry, right? Sincere ... Meanwhile, you can quite easily make such a request using special characters in the search bar, which are called the query language.

1. Exclusion / inclusion of certain words - signs “+” and “-”

Suppose you want to find sites that have jokes about blondes, but you don’t endure jokes in which evil blonde people call fools. Then in the request you write the following: “+ joke + blonde-fool” - this means a team to search for pages that necessarily contain the words “joke” and “blonde”, but the word “fool” is missing.

2. Enumeration of alternatives - sign "|"

The vertical bar allows you to set alternatives: the system searches for at least one of the listed words. For example, if you want to find pages where one of the words “dad”, “mom”, “daughter”, “granddaughter” appears, the search query will look like this: “dad | mom | daughter | granddaughter".

3. Search for an exact match - a “!”

Typically, search engines look for all word forms of the entered word, even if it is fully specified: for example, if “blonde” is entered in the line, then the pages where exactly “blonde” is found will be displayed first, but then in the search results will appear “blondes”, “blonde” , “Blonde,” “blonde,” and so on. However, if you need to find only this specific form - “blonde” - then enter the search bar like this: “! Blonde”.

4. Search for the exact phrase - quotation marks

If you need the exact phrase “hypervisor quasiconverter”, and not “hypervisor analyzer that includes a dark green quasiconverter in pimples”, then when searching, enclose the phrase in quotation marks: “hypervisor quasiconverter”, in which case the search engine will only display those pages where these words are strictly next to each other.

5. Setting the distance between words - “/ n”

In the case when you want to find all variants of the phrase “tall [any word] blonde” - you can reflect this irrepressible desire in the search query using the “/” sign, meaning “did not exceed”, and a number showing the allowable number of words, and 1 - it means the absence of words. I explain. The query “high / + 2 blonde” means a team to search for all combinations where between “high” and “blonde” there is no more than 1 word. That is, a “tall blonde” - will do, “a tall long-legged blonde” - will do, but a “tall long-legged and passionate blonde" will no longer exist. The “+” sign means that the word (s) should be on the right, and “-“ - on the left. That is, the query “high / (- 2 4) blonde” means that “high” should be from “blonde” in the range of distances from 2 words on the left to 4 words on the right.

6. Address restriction - the "#url" command

If you are interested in searching, for example, only on a specific server or by a specific group of addresses, or, conversely, searching that excludes specific addresses or groups of addresses, you can use the command # url \u003d "address or group". For example, if you want to find all pages on which the word “quadroid” appears, excluding the site www.kvadroid.ru, give the following command: “quadroid ~~ # url \u003d" www.kvadroid.ru "."

7. Search for links - the "#link" command

Sometimes (this is especially true for owners of their own pages) it can be very interesting to find out which sites link (put a hyperlink) to a particular address. In this case, it is enough to specify the following data in the search bar: "# link \u003d" address "". That is, the command "# link \u003d" site "" will show all the pages where links to

I have listed only the main features of the search engine query language Yandex. I note that these parameters, as a rule, are very similar in other search engines, although in any case it is advisable to look at the description on a specific search engine before using them.

It should also be borne in mind that almost every search engine allows you to create fairly complex queries using a special form called “advanced search”, for which you no longer need to know how the query language works in this search engine.

In order to successfully promote your site, promote it in search engines, you must have at least a general idea of \u200b\u200bthe work search engines.

There are many billions of web pages (documents) on the Internet at the same time, and this number is growing every day. It is clear that in such a huge amount to find exactly what you need, an incredibly difficult task. Moreover, without special services - search engines, or machines - this task is practically unsolvable.

How, when and which spider will come to your site is the concern of the search engine itself. We can influence only indirectly, through frequent publication. But no matter how high the frequency of publications on your site, you can never compare with information resources (such as ITAR-TASS) or social networks (Facebook, etc.), where search engines literally live.

But it can be firmly asserted that the less often you publish new content on your site, the less often Spider will come to you.

We can and must take this moment into account in our work to resolve certain issues. It is clear that a link from a social network (especially owned by the search engine itself - Google+ or Ya.ru) will be the most quickly detected by the search engine robot, but it will not be able to transfer the same weight as the link from the most authoritative resource on your topic.

Therefore, if you need your site (or its page) to find search engines as quickly as possible, then you need to get links from information resources, social networks, etc. And if you are concerned about "puzomerki" (PR and TIC), then you need completely different links. In order for this to become clearer to you and you did not make strategic (that is exactly so!) Mistakes, we need to find out how search engines deliver and generally rank pages.

Conclusion: search engines are unusually complex and closed structure, but the webmaster needs to know the basic principles of its work

P.S. Knowing these basics will help you promote any site, for example, a site whose theme is ferrosilicon. And even if you in the ferrosilicon do not really understand. But you will be well versed in SEO.

Do you like the article? Share with friends: