Technology development trend of search engine

Intermediary transaction http://www.aliyun.com/zixun/aggregation/6858.html ">seo diagnose Taobao guest cloud host technology Hall

Search engine after several years of development and groping, more and more close to people's needs, search engine technology has also been a great development. The latest technological developments in search engines include the following:

First, to improve the search engine on user retrieval questions to understand

In order to improve the search engine's understanding of user retrieval questions, it is necessary to have a good retrieval language, in order to overcome the shortcomings of keyword search and directory query, now there is a natural language intelligent answer. Users can enter simple interrogative sentences, such as "How can kill virus of computer?". Search engine after the structure and content of the analysis of questions, or directly give answers to questions, or to guide users from a number of alternative issues to choose from. The advantage of natural language is to make the network communication more humane, and to make the query more convenient, direct and effective. In terms of the above example, if the keyword query, most people will use the word "virus" to retrieve, the results will inevitably include various types of virus, how the virus is produced, and many other invalid information, and with "How can kill virus of computer?", Search engine will be how to kill the virus information provided to users, improve the retrieval efficiency.

Ii. processing of the results of the search

1 search engine based on link evaluation

The excellent representative of search engine based on link evaluation is Google, and its original "link evaluation system" is based on the understanding that the importance of a Web page depends on the number of links to other pages, especially the number of pages that have been identified as "important". This evaluation system is very similar to the "sci-tech citation index", but because the Internet is developed in a commercial environment, the number of links to a website is closely related to its business promotion, so the evaluation system lacks objectivity to some extent.

2) based on access to popular search engines

The representative of a search engine based on access to mass is direct hit, whose basic idea is that most of the sites that most people choose to visit are the most important. According to the previous tens of thousands of web users in the search results are actually selected and visited the site and their time spent on these sites to determine the importance of the site rankings, and to determine which sites best meet the user's search requirements. Therefore, it has a typical characteristic of chemotaxis. This evaluation system has the same disadvantage as the search engine based on link evaluation.

3 Remove additional information from the search results

Some surveys have pointed out that too much additional information to increase the user's burden of information, in order to remove these excessive additional information, you can use user customization, content filtering and other search techniques.

Third, determine the search engine information collection scope, improve the search engine targeted

1 Vertical Theme Search engine

Information on the Internet, network resources to 10 times times the growth of the speed, a search engine is difficult to collect all the network information on all topics, even if the information topics collected more comprehensive, because the topic scope is too wide, it is difficult to all the topics are accurate and professional, so that the results of garbage too much. In this way, the vertical theme of the search engine with its high degree of targeting and specialization in various types of search engines occupy a place, such as stocks, weather, news and other categories of search engines, with high pertinence, users of the query results of higher satisfaction. The author thinks that the vertical theme has great development space.

2) Non-WWW information search

Provide retrieval of class information such as FTP

3 Multimedia Search engine

Multimedia retrieval mainly includes sound and image retrieval.

Four, the search engine technology development focus on the search results processing, provide more optimized search results

1 Pure Search Engine

This kind of search engine does not have its own information collection system, uses other people's existing index database, mainly pays attention to the retrieval idea, the technology and the mechanism and so on.

2) Meta search engine

Now there are many search engines, the scope of their collection of information, search mechanisms, algorithms, etc. are different, users have to learn the use of multiple search engines. Each search engine can only relate to the 30–50% of the entire WWW resource (search engine watch data), which results in less than 34% repetition rate of the query result obtained by the same search engine, and less than 45% of each search engine.

Meta search engine (metasearch enging) is the user submitted retrieval request to a number of independent search engines, and the results are centralized unified processing, in a unified format to provide users, so there are search engines above the search engine. Its main focus is to improve search speed, intelligent processing of search results, personalized search function settings and user search interface friendliness, recall and precision ratio are relatively high. At present, the more successful meta search engine has MetaCrawler, Dopile, Ixquick and so on.