基于Wiki资源的中英文跨语言本体知识库构建

随着互联网的快速发展，原来由英文信息占主导地位的Web已经发展成为全球化、多语言的信息空间。多语言的Web环境为语义信息交互与共享带来了新的挑战。构建大规模的跨语言本体知识库对于实现计算机跨语言信息的自动处理有着十分重要的作用。现有的跨语言知识库存在不同语言之间实例数量不平衡、跨语言链接少和中文结构化信息贫乏的缺点。针对这些问题，本课题以基于Wiki资源构建中英文跨语言本体知识库为目标，拟从四个方面展开研究：（1）异构中文Wiki资源的匹配与合并；（2）中英文Wiki资源的跨语言链接发现；（3）中英文跨语言本体自动构建；（4）面向Wiki资源的跨语言结构化信息抽取。基于上述研究获得的中英文跨语言本体知识库，可为中英文跨语言信息共享与处理提供有效支持。

With the rapid development of Internet,the Web has evolved into a glabal multilingual information space. The multilingual nature of Web poses new challenges for semantic information interaction and sharing. Building large-scale crosslingual ontology knowledge bases is crucial for automaticly crosslingual information processing by computers. Existing crosslingual knowledge bases suffer some problems, including imbalanced number of instances among different languages, small number of crosslingual links, and insufficient Chinese stractural information.To solve these problems, this project aims to build large scale Chinese-English crosslingual ontology knowledge base from Wiki resources.Specifically, four topics are to be investigated in this project: (1) heterogeneous Chinese Wiki resources matching and alignment; (2) crosslingual links discovering across the Wiki resources; (3) crosslingaul ontology construction; (4) crosslingual information extraction from Wiki resources. Based on the above research topics, a large scale Chinese-English crosslingual knowledge base can be built, which may greatly benifit corsslingual information proccessing applications.