Network Search Engine and Intelligent Agent Technology
Abstract: Analyze the principles of search engines and intelligent agency technology, discussing the importance of both the present and unable to network information retrieval. [Key words] intelligent agent search engine intelligent information retrieval
The rapid development and maturity of the Internet will promote the rapid expansion of information in various fields of society, to find a rich information source for people, but also provide a challenge to the accurate positioning of the information. Providing online resources is one of the important contents of network information services, and modern people also put forward increasingly high requirements for the correctness and comprehensiveness of information grasping, and therefore, it is a matter of urgent network information retrieval tool.
In 1993, after the birth of the first search engine, it has gradually been mature, accompanied by the continuous development of computer intelligent research, intelligent agent technology with adaptability and learning characteristics is also transitioning from the experimental phase to the actual application. Currently, Search Engine and Intelligent Agent have become a key technology and core idea of network information search.
1 Development status of search engine technology
1.1 Search Engine Technology
At present, the main network information retrieval technology is the search engine technology. The search engine is actually a dedicated WWW server, or it is a class of websites on the Internet. This type of website is different from the general website, its main work is to collect Thousands of websites and web information on the network make up the huge index database. Use excellent search engines to achieve a half-time effect. At present, there are about more than 3,000 search engines on the network. We are more familiar with SINA, SOHU, YAHOO, NETEASE, and Chinese Excite, etc.
In general, search engines mainly take two ways to retrieve network information resources, one is to use the classification topic directory, the website is classified, and the link to the website must at least belong to one of the categories, forming a similar book The same classification topic catalog is browsed by step by step by step by step by step by step by step by step, and the search engine that uses this search method is Yahoo, Sohu et al., Due to the use of experts to summarize and classify, bring information navigation It is very convenient, but this method requires a lot of manpower in the classification and directory finishing; the other is to use keyword matching methods, and its processing object is mainly text, it can establish a large number of documents to create a word (word) to the document. Index library, on this, the user uses keywords to search for web pages, and the system will display all websites, web pages, and news, including the search terms, web pages, and news. Keyword search can solve the search problem for web pages. As long as the user enters keywords, the system automatically retrieves within the selected range through the spider robot, and the searchable information automatic marking into the index database, match the detected The web page in the range can be obtained.
1.2 Information retrieval technology used by search engines and its shortcomings
The information retrieval technology used in the current search engine is: Robot technology, indexing technology, translation technology, conversion technology, filtering technology, database technology, result processing technology, etc. The biggest advantage of the search engine is that the coverage of the information is large, the information is novel, and the search engine will be arranged before the search results of the search results are arranged before. However, due to the restrictions on the intelligent level of information retrieval technology used by the search engine, and there are many deficiencies in the search for network information. There is mainly aspects of the following aspects.
(1) The current search engine is mainly downloaded to the self-built index library through Robot, because many downloaded pages are useless or temporary information, which affects the retrieval speed and increases the user retrieval burden.
(2) Since the search engine generally uses a keyword search method, in many cases, the user is difficult to accurately express the real needs of information, and difficult difficulties have caused difficulties in retrieving difficulties with keywords or keywords. .
(3) The coverage of each engine is quite limited. After investigation, it was found that there was no search engine with an index of more than 1/6 of the entire web page. (4) The result of the search is not accurate. The accuracy of search results is determined by the correlation of query words and web pages, often entering a single query word to return tens of thousands of results, or zero results.
2 Intelligent Agent Technology
2.1 Intelligent Agent
Intelligent agents, also known as intelligent, is a new achievement of artificial intelligence research. It is in the case where the user does not clear and specific requirements, according to user needs, instead of users to carry out various complex work, such as information query, filtering and management, and can Specifies the user's intention, self-developed, adjusts and implements the work plan. It is intelligent and is a proxy software that can be advanced, complex automatically. Intelligent agents can be applied to a wide range, a hot spot in the artificial intelligence area in recent years, and is used in the field of information retrieval, becoming one of the important technologies for development intelligence and personalized information retrieval.
2.2 Characteristics of Intelligent Agents
1 Intelligence. With rich knowledge and certain reasoning ability, it can speculate user intention, and can deal with complex difficult tasks. The needs of users can be analyzed, automatically reject some unreasonable or may give users hazard requirements. And have the ability to learn from experience, proper self-adjustment, improve processing problems.
2 agent. It is functionally a user of the user, which can do some tasks in place of the user, and take the initiative to feed back to the user.
3 mobility. You can travel to any target host on the network and perform information processing operations on the target host, and finally return the result set to the starting point, and can move with the mobile user.
4 Active. Depending on the user's needs and environmental changes, actively report and provide services to users.
5 collaboration. Information exchange can be communicated through various communication protocols and other intelligent body, and can coordinate complex tasks with each other.
3 Combination of search engine technology and intelligent agency technology
Search engine and intelligence agent technology have their own advantages and shortcomings, combining these two technologies, will provide a broad world to develop a new generation of more powerful online information search systems. Intelligent agent mainly integrates special environments, with user interest to complete search. It identifies user information, preferences, and summarizes, analyzes users' interest hobbies, and automatically, automatically and independently uses users to find their interested information. Combining search engines with intelligent proxy technology is an inevitable trend of establishing a new search mode.
3.1 Introduction of personalized services in server-side
The idea of absorbing intelligent agent technology is absorbed by the server side, introducing the idea of personalized and humanized services. The introduction of the user feedback mechanism to improve the retrieval mechanism, improve the search hit rate, and can also provide personal special retrieval services. This approach can be implemented in the form of an account, which provides an account for each user (similar to the personal mailbox) to record the user query trace, so that when the user logs in again, it combines the conventional user search record to cooperate with the conventional user retrieval record. Search service. This model reflects the characteristics of personalized services. For information on the user's consistent query, you can directly extract from the user's information base to avoid repeated queries. In addition, by tracking the feedback of the user, obtain the evaluation of the user's results, thereby increasing the retrieval quality. The natural linguistic development of the retrieval entry will help optimize the search interface and improve the user-friendly interface.
3.2 Expansion of client intelligent agency technology
The intelligent search agent technology is mainly, combined with the search engine "theme" retrieval mode, pay attention to individual needs, improve information and user needs related systems, and exchange information, exchange information, exchange information through unified transport protocols, so that more Many information can be excavated to compensate for the limited defects for smart agent information. This model makes full utilize the liquidity, interactivity, intelligence characteristics of the intelligent search agent, while absorbing the idea of search engines, providing new modes for high quality information personalized retrieval services.
