SBIR Phase I: Quality-Based Knowledge Discovery for Information Retrieval in Large Organizations

SBIR 第一阶段：大型组织中基于质量的信息检索知识发现

• 批准号: 0232777
• 负责人: Gary Robinson
• 金额: $ 9.98万
• 依托单位: Transpose, LLC
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2003
• 资助国家: 美国
• 起止时间: 2003-01-01 至 2003-12-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0232777&HistoricalAwards=false
• 关键词: SBIR; Phase; Quality; Based; Knowledge

This Small Business Innovation Research Phase I project will determine the feasibility of a novel, powerful technique to determine the quality of documents and the reliability of human reviewers in a large community, such as a company, government agency (particularly intelligence organizations), or university. Many organizations have difficulty locating their own knowledge. They cannot determine the quality of their internal documents nor the ability of their employees to evaluate those documents. Important information may get buried in layers of bureaucracy. This research seeks to address these problems by measuring document quality and user reputation, based on human evaluation and automated processing of those evaluations. The research objectives are: study and refinement of information theoretic techniques underlying the current technology; discovery of practical designs for expanding its scope; and creation of proof-of-concept demonstrations. The research will produce technical papers describing such systems and prove them feasible through software. Potential applications include stand-alone knowledge management products focused on document quality and plug-ins to add this capability to existing KM systems. Members of communities using these products will find the information they need more quickly and with greater confidence, enhancing efficiency and effectiveness in corporations, academic institutions, and government agencies.The research will lead to marketable knowledge management software products and/or services which offer quality-based knowledge discovery and are targeted to large organizations in any sector (business, public, academic, or not-for-profit). This technology is not another kind of search engine. It is an evaluation system that identifies the best sources of knowledge, and the most valuable knowledge items, based on the experience of the users in the community. Today there are many vendors providing a variety of knowledge management solutions to large organizations, but we know of none that offers a knowledge discovery solution based on quality.

这个小型企业创新研究第一阶段项目将确定一种新的、强大的技术的可行性，以确定文件的质量和大型社区(如公司、政府机构(特别是情报组织)或大学)中人类审查员的可靠性。许多组织很难找到自己的知识。他们无法确定其内部文件的质量，也无法确定其员工评估这些文件的能力。重要信息可能会被层层官僚机构掩埋。这项研究试图通过基于人工评估和对这些评估的自动处理来衡量文档质量和用户声誉来解决这些问题。研究目标是：研究和改进作为当前技术基础的信息理论技术；发现扩大其范围的实用设计；以及创建概念验证演示。这项研究将编写描述这种系统的技术论文，并通过软件证明它们是可行的。潜在的应用包括侧重于文件质量的独立知识管理产品和将这一能力添加到现有知识管理系统的插件。使用这些产品的社区成员将更快、更有信心地找到他们需要的信息，从而提高企业、学术机构和政府机构的效率和效力。研究将导致可销售的知识管理软件产品和/或服务，这些产品和/或服务提供基于质量的知识发现，面向任何部门的大型组织(商业、公共、学术或非营利组织)。这项技术不是另一种搜索引擎。它是一个评估系统，根据社区用户的经验确定最好的知识来源和最有价值的知识项目。今天，有许多供应商为大型组织提供各种知识管理解决方案，但据我们所知，没有一家提供基于质量的知识发现解决方案。