CIF: Small: A Theoretical Framework for Dynamic Collaborative Online Information Searching

今天的互联网已经从一个提供连接的网络转变为一个巨大的人类和机器知识库，与人类生活几乎每个方面相关的信息都存储在某个角落。搜索引擎允许基于关键字搜索这些知识，虽然自然语言查询越来越有用，但搜索复杂信息通常需要人类思维（通过搜索引擎的能力增强）才能获得有用的搜索结果。日常问题需要个人在互联网上搜索相关的想法，这一任务可能会因缺乏适当主题方面的专门知识而受到阻碍。随着普遍的在线连接，有可能无限的“专家”（或只是助手）在线贡献他们的知识，与他们的专业知识有关的问题。该项目设想利用这些潜在的专家（或助手）来处理大量的信息请求，利用在线协作的力量。随着世界走向知识型经济，利用因特网进行协作，其中很大一部分涉及搜索信息，这一项目的相关性应日益明显。该项目将培养学生谁将理解和扩展所提出的方法。本计画提出一个新的理论架构来分析资讯搜寻的动态性。它包括处理多个请求的信息搜索，来自不同领域的多个专家的协作，探索专业知识与生成的请求的人群不匹配的概念，以及专业知识的多样性，以便更快地搜索多学科请求。虽然要研究的问题与多类排队和调度有相似之处，但信息搜索独特地使用了信息论的思想。这些研究活动需要信息论、排队论、调度和社会网络概念等领域之间有趣的相互作用。该奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

The Internet today has been transformed from a network providing connectivity, to a massive repository of human and machine knowledge, with information relevant to nearly every aspect of human life stored in some corner. Search engines allow keyword-based search of this knowledge, and while natural language queries are increasingly useful, searching for complex information often requires human thinking (augmented with the capabilities of search engines) to obtain useful search results. Routine problems require individuals to search the Internet for relevant ideas, a task which may be hindered by lack of expertise in the appropriate subject matters. With pervasive online connectivity, there are potentially unlimited ‘experts’ (or just helpers) available online to contribute their knowledge to problems related to their expertise. This project envisions tapping such potential experts (or helpers) to handle a large number of requests for information, leveraging the power of online collaboration. The relevance of this project should be increasingly apparent as the world moves towards a knowledge-based economy where the Internet is used for collaborative work, a significant part of which involves searching for information. The project will train students who will understand and extend the proposed approach. This project proposes a novel theoretical framework to analyze the dynamics of information searching. It includes information searching to handle multiple requests, collaboration of multiple experts from different fields, exploration of the concepts of expertise mismatch with the population of generated requests, and of diversity of expertise to allow faster searches for multi-disciplinary requests. While the issues to be studied bear similarity with multi-class queuing and scheduling, information searching uniquely uses information-theoretic ideas. These research activities require an interesting interplay between the fields of information theory, queuing theory, scheduling, and social network concepts.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.