This is an article published by Google Engineer Amit Singhal on Google's official blog, which describes some of the technologies behind Google's search rankings, including Google's understanding of webpages, semantics, and user intentions.

Source: Internet
Author: User
Xeim this is an article published by Google Engineer Amit Singhal on Google's official blog, which describes some of the technologies behind Google's search rankings. It involves Google's understanding of webpages, semantics, and user intentions.
The core technology of Google search ranking is derived from information retrieval (IR), an academic topic that has been 50 years old. IR uses statistical principles to study attributes such as the frequency of use of text and rank the results. Google search, based on IR theory, uses the link, webpage structure, and other technologies to form a unique search technology.

Understanding web page:

Google has invested heavily in web crawlers and indexing systems over the years. Therefore, Google has a very large and up-to-date web index. In addition, Google also uses some of the latest technologies to improve the indexing quality. For example, they have developed a technology that can understand the important concepts expressed on a web page literally. people search for "Galleria Sprovieri londra" in the Italian language and will find Sprovieri Gallery in London, although the Sprovieri gallery home page does not contain either London or londra. In the United States, people search for "Cool tech PC Vancouver, wa" and will find However, there is no text on the homepage of indicating that they are located in Vancouver. Other technologies include distinguishing important or non-important words in a webpage and the freshness of the webpage content.


Google can use several search keywords provided by users to understand users' real intentions. They are very advanced in terms of spelling correction, meaning, and conceptual analysis. Many people have more or less experienced Google's spelling correction function, such as searching for "kow.annan". Google will ask if you are searching"KofiAnnan ", however, when someone searches for" kow.beans ", Google will correct it to" coffee beans ". (Google is actually trying semantic technology.-translator)

The meaning of words is the foundation for Google to try to understand the query semantics, and is also the biggest difficulty Google has encountered. Some things are obvious to humans, but it is difficult for machines to handle them automatically. Users do not want to query the words they use. People do not even know what to use for queries. In this case, Google's word meaning system can play a role, and the word meaning system can make very complex corrections to query statements. For example, when querying "Dr Zhivago", Google knows the representative of dr.DoctorWhen querying "Rodeo Dr", Dr represents drive. When you search for "back bumper Repair", the result is:RearBumper repair, and search for "Ramstein AB", Google can understand AB as Air Base, "B & B AB" will understandBed and breakfasts of Alberta. Google has developed this semantic understanding system into hundreds of different languages.

Another technology used by Google in search rankings is concept recognition, which can identify the queried content. For example, we can query "New York Times Square Church ", google knows that we actually look at the famous church in Times Square in New York, rather than an article in The New York Times. Concept recognition technology is more than that. Google also strengthens it to correctly identify semantics. For example, searching for "PC and its impact on people" is actually the influence of search computers on society. Google's search and analysis algorithms are widely used and target almost all languages.

Understand users:

Google tries to understand that users aim to return the results they really need for users, rather than what they say in the search statement. This technology is based on a world-class localization system, along with advanced personalization technologies and a variety of user intention recognition technologies.

Google's emphasis on local results is reflected in their localization work. The same query statement returns different results in different countries. For example, if the query is "Bank]", a bank is returned in the United States, and a bank fashion clothing chain store may be in the UK, or a bank in the UK, and in other English-speaking countries, such as Australia, Canada, New Zealand, and South Africa, the return should still be a local bank. If you query this word in some non-English countries, such as Egypt, Israel, Japan, Russia, Saudi Arabia, and Switzerland, the returned results will be more interesting. Similar to football indicating different sports in the United States and Britain, the query results of the same word in different countries may be quite different.

Personalized query is another advanced search technology of Google. If a user who has logged on has activated the web history service, as the query time increases, Google will query the history based on his query history, automatically adjust the returned results. For example, a person who frequently queries football-related topics will gradually get more football-related results from Google. If you prefer the results returned by a shopping site, more results will be obtained from the shopping site in future queries.

Another example of Google's return to the user's desired results is that if you search for "Chevrolet Magnum", we know that Magnum is not produced by Chevrolet, but by Dodge, Google will automatically returnDodgeThe result of Magnum. In another example, someone searches for "Bangalore" and returns not only the home page of the city of Bangalore, but also a map of Bangalore and Videos related to the city and traffic of Bangalore, these videos will make you feel immersive.

International Source:
Comsharp CMS

Contact Us

The content source of this page is from Internet, which doesn't represent Alibaba Cloud's opinion; products and services mentioned on that page don't have any relationship with Alibaba Cloud. If the content of the page makes you feel confusing, please write us an email, we will handle the problem within 5 days after receiving your email.

If you find any instances of plagiarism from the community, please send an email to: and provide relevant evidence. A staff member will contact you within 5 working days.

A Free Trial That Lets You Build Big!

Start building with 50+ products and up to 12 months usage for Elastic Compute Service

  • Sales Support

    1 on 1 presale consultation

  • After-Sales Support

    24/7 Technical Support 6 Free Tickets per Quarter Faster Response

  • Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.