• Home
• About

At the beginning was the Word, and the word that was on the site, and the word that was written in hypertext markup language. And the language that was so easy and simple, that the number of words on the Internet is growing exponentially. And urgently needed mechanism that can find exactly what word you want, and these mechanisms have been search engines. In 1993, launched the first search engine «Aliweb», which is working now, but enjoy it, or aristocrats, or degenerate. In 1994, almost immediately after the massive spread of the Internet has created the first known search engine «Yahoo!». The founders and developers of the first search engine used the method of finding books in libraries. They created a thematic catalogs, categories, which

were each site. A man came into the directory, select the appropriate headings and got several sites on topics related to it. In the beginning, when the site was not much, everything was fine, but then was getting harder and harder to find the desired resource. Topic has become more and more sub-categories all grew and grew. In the end all to the group of «Scooter» had done in the 7-8 sub-path, and if in any category of «turn» is not there, it was just nothing to find. Then gradually all the search engines have begun to move on «search indexes». Unlike directories, search indexes form ligaments «Request - reply», and «replies» attributed to several different resources. But later it turned out that it is difficult not only for humans but also for search engines, because any common request «empty» hundreds of thousands of responses, which simply could not understand. And that is why the choice of search engine that «mountains» of useful information twenty-thirty and became a defining links in the search market. Most search engines switched to a four-structure work. The first three stages of almost all search engines are identical. First, the search engine indexes (collect) the information and put them into a database, and then, taking into account the morphology, cuts all the words on the page elements. If a person is drawn to the search processor, the operation enters the third and fourth stages of the search, she selects from the database are all the documents that match a given search and then displays them in depending on which page has more search weight, accuracy of re-entry search phrase, date, updates, etc. Such a sample, based on many parameters, called the ranking of search results. The first mechanism for ranking documents based on the basis of the accuracy of entering the text in the search query. But this proved irrelevant because the extent of entering a search phrase, its location in the body, or a list of keywords does not give a complete picture of the document. The most famous class of formulas for calculating the weight of the document in relation to the inquiry, called TF * IDF. TF * IDF - the numerical measure of conformity and word document, but more so, than the relatively more frequent (TF - the term frequency in document), the word in the document and has received relatively less (IDF - rare words; value, the inverse frequency words in the collection) in the collection. That mechanisms for ranking search engines differ from one another. Now, each search engine tries to offer their tools for ranking documents in order to take a leading position in the search market. Many search engines search results to affect approximately 200-300 of factors, which are carefully kept away from competitors. It is they (together with the speed of indexing and volume indexes) and helped in their time ahead of Google once the leading Altavista.

Comments

No comments yet.

Leave a comment

Name (required)

E-mail (required)

URI

Your Comment