CAREER: User-centered Adaptive Information Retrieval

信息检索（Information Retrieval， IR）是指从大量文本中查找信息，是克服信息过载最有用的技术之一。例如，网络搜索引擎现在是每个人在网络上查找信息的基本工具。现有的IR系统本质上不是最优的，因为检索决策主要基于当前查询和文档集合，而不考虑有关用户和搜索上下文的信息。本项目旨在消除现有检索方法的这一限制，并正式开发一种新的检索范式，称为以用户为中心的自适应信息检索（UCAIR），其中用户信息和搜索上下文都被利用来提高检索性能。本课题主要研究内容包括：(1)基于贝叶斯决策理论开发新的UCAIR框架；(2)开发新的语言模型，利用用户信息和搜索上下文来提高检索精度；(3)开发新的检索方法，优化整个检索过程的长期检索效用；(4)开发新的检索方法，利用用户相似度，根据其他类似用户的信息更好地推断特定用户的信息需求；(5)开发用于网络和生物信息学文献检索的UCAIR原型系统。该项目将通过为UCAIR开发一个统一的正式框架、利用用户信息和搜索上下文的各种检索方法来提高检索精度，从而推动信息检索技术的发展，这将直接导致在所有领域更有效的信息检索应用。研究结果亦有助加强现行资讯科技课程，提高资讯科技人才的教育水平。项目网站http://sifaka.cs.uiuc.edu/ucair/将用于传播研究成果。

Information Retrieval (IR) refers to finding information from large amounts of text, and is among the most useful technologies for overcoming information overload. For example, Web search engines are now essential tools for everyone to find information on the Web. Existing IR systems are inherently non-optimal because the retrieval decision is made primarily based on the current query and the document collection without considering information about the user and search context. This project seeks to eliminate this limitation of the existing retrieval methods and formally develop a new retrieval paradigm called user-centered adaptive information retrieval (UCAIR), in which user information and search context are both exploited to improve retrieval performance. This project includes research on: (1) developing a new UCAIR framework based on Bayesian decision theory; (2) developing new language models to exploit user information and search context to improve retrieval accuracy; (3) developing new retrieval methods to optimize the long-term retrieval utility over an entire retrieval session; (4) developing new retrieval methods to leverage user similarities to better infer one particular user's information need based on information about other similar users; and (5) developing prototype UCAIR systems for searching the Web and bioinformatics literature. The project will advance the state of the art of IR through developing a unified formal framework for UCAIR, a variety of retrieval methods for exploiting user information and search context to improve retrieval accuracy, which will directly lead to more effective information retrieval applications in all domains. The research results will also enhance the current IR curricula, improving education of information technology workforce. The project Web site http://sifaka.cs.uiuc.edu/ucair/ will be used for research results dissemination.