Information Retrieval and Mining in Web Archives

网页最重要的特征之一是它们能够及时改变内容和结构。许多流行的网页不断变化、发展并提供新的信息。最近，互联网社区认识到需要存储和保存过去的网络内容，以便将来可以使用。本研究旨在利用网页档案来推断网页的时间特征，进行知识发现，并改善用户在网页上的体验。从网络档案中搜索、浏览和发现知识的最新技术仍然相当有限，用户很难使用。其主要目标之一是利用隐藏在网络或新闻档案中的信息来改进网络搜索、浏览和其他常见任务。

One of the most important characteristics of web pages is their capability to change content and structure in time. Many popular web pages continuously change, evolve and provide new information. Recently, Internet community recognized the need to store and preserve past content of the Web so it could be available for future use. This research project aims at utilizing web archives for inferring knowledge about temporal characteristics of web pages, for knowledge discovery and for improving user experience on the Web. State-of-the-art techniques for search, browsing and knowledge discovery from web archives are still rather limited and difficult for users to use. One of main objectives is to use information hidden in web or news archives for improving Web search, browsing and other common tasks.