| With the development of information technology and the popularization of Internet, Web has already become an important way that people obtain information. Obtaining various information from the Internet has been an important part of people's daily life. In recent years, some people particularly concern some accident news, such as mine explosion, plane crash, terrorist attack, ect. The government needs to master the accident news taking place both at home and abroad accurately and promptly, and then offer scientific basis to take precautionary measures effectively in time. How to search useful information from a large number of information and obtain the leitmotiv of information, and how to read the new information quickly have become a problem which needs to solve. How to find out and use information that users need has been the focus of academe. According to users' interests and visiting mode, recommending information actively to users(individual service) has already become a hotspot. However, solving these problems needs to research the resource to be dealt with. At present, the resources dealt with by individual recommending system are all categories of texts, so some kinds of technology that deal with text get wildly promotion and development. These input the energy for computational linguistics, and have closely relation with analysis of text structure. Analysis of text structure is the foundation of technology. The purpose of this article is giving a structure of accident news based on concept on the basis of analyzing accident news and text characteristics. This structure analyze text information from semantic meaning, and offer a little basis for individual service. All the work and contributions of this article summarize as follows:1. Analyze the characteristics of the accident news , including characteristics of words, the usage of concurrently kind of words, matching characteristics, and provides the detailed instance to every kind of characteristics. 2. Use Hownet to map concept to accident news that have been carved and labeled, and divide the hierarchical of accident news through concept mapping.3. Give a form of accident news that have been divided levels based on concept and distill abstract based on statistic. |
PDF Full Text Request