Natural language processing is to study how to make computers understand and generate people's daily (such as Chinese, English) language, making computer understand the meaning of natural language, and the problems proposed by people, through the way of dialogue, with natural language Reply. The purpose is to establish a close and friendly relationship between a person and the machine, making it a high degree of information transfer and cognitive activities.
The natural language understanding system can be used as an expert system, knowledge engineering, intelligence retrieval, and office automation natural language human interface, there is a lot of practical value.
The research of natural language treatment mainly has two ideas of "rationalism" and "empirical". The representative technique of "rationalism" is the syntax of Jumsky's conversion generating syntax. The core is phrase structure rule. The process of analyzing the structure structure is the use of rules from top or bottom-up syntax generation process. . On this basis, the researchers have successively proposed a semantic representation of semantic networks, CD theory, frame frameworks. These grammar and semantic theories have gradually begun to combine each other. And derive a lot of new syntax theories, representative of the word function syntax (LFG), functionality, and a general phrase structure syntax (GPSG).
The "rationalism" is relative to "empiric" research ideas, mainly to research on large-scale spending. The corpus is a collection of a large number of texts, and the knowledge provided is represented by statistical strength, not certain, with scale, is intended to cover a comprehensive language.
At the same time, with the rapid development of the Internet and broadcases, online information explosion growth. How to get valuable information on huge Internet has become a problem that users are increasingly concerned. Therefore, in a certain strategy to collect, discovery information, understand, extract, organize, and process information, and provide users with search services to provide users with the purpose of information navigation. Search technology. Search needs. On the current common search engine still have a lot of limitations, such as information loss, return information is too much, information has nothing to do, which makes users still dissatisfied with existing search technologies, and look forward to the emergence of more perfect search technology.
Due to the three functions of natural language understanding, that is, machine translation, semantic understanding, and human chance technology can give search technology more humanized and easy to use.
The natural language understanding technology in the search engine is currently in the search engine is the machine translation and semantic understanding technology. The search engine applied to these technologies is called a smart search engine. Since it has improved information, it has certain understanding and processing capabilities of knowledge (or concept) from the current keyword level, and there is a certain understanding and processing capability of knowledge, thereby intelligent, humanized characteristics. It allows users to use natural language to search for information, providing them with more convenient and more exact search services.
(Analysis of the research significance of the paper or engineering project, analysis of domestic and foreign research status, with main references)
The search engine is a tool that is widely used by the Internet of STA in addition to email. The traditional search engine is a keyword as the retrieval method, often has a big error in the accuracy, bringing a lot of disambiguation results, and the user has to spend a lot of time for retrieval.
After using advanced natural language understanding, the search engine can identify and answer users' questions, and use users to get rid of the traditional search engine based on keywords, guide users to find the required information more efficiently and more quickly, at the same time for users Provide relevant content with reference value. Due to these features, smart search technology can be widely available in all aspects of Internet information retrieval. It can provide background support for large integrated search engines that make it characterized by humanized and interactive. It can easily search for the professional category of vertical search engines. Of course, it can also provide information on the information portal to provide a convenient and fast station information search service.
In foreign countries, ask.com has created a search method of natural language understanding, which is welcomed by users. The domestic natural language understanding technology as the core is also intelligent search engine in China. This search engine can fully understand the search for search engines than previously keyword-based search methods. The search results are more accurate, correlation, and practicality, greatly saving search time. At the same time, this new search engine also has a learning function, and its technology can track and fully analyze all the questions that the searchers have emitted, so that similar questions will have a better answer. In addition, in the field of business intelligence (BI: BusinessIntellignce) and enterprise applications, there is a large amount of data information needs to be handled, how to manage these data more efficiently, how to find the information you need in time, and the common problem in this area. Therefore, developing a search engine with natural language processing technology, not just for the portal, but also for enterprise applications, as a valid tool for their content management and information retrieval, providing a powerful assistant for online e-commerce such as B2B / B2C, It has a practical meaning.
Currently, companies such as www.kanisa.com provide companies based on the user-centric content management system based on natural language, which provides Site query features such as Microsoft, Novell, eBay, Gateway.
