Theory of multi-level electoral college for multi-candidte elections and electoral college based face recognition surveillance and intelligent textual information retrival system

多候选人选举的多级选举团理论及基于选举团的人脸识别监控和智能文本信息检索系统

• 批准号: 261403-2006
• 负责人: Chen, Liang
• 金额: $ 1.31万
• 依托单位: University of Northern British Columbia
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2007
• 资助国家: 加拿大
• 起止时间: 2007-01-01 至 2008-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=363654
• 关键词: Theory; multi; level; electoral; college

On the theoretical front, we will study the stability of the multi-level Electoral College voting system for multi-candidate elections. Its simplest form, the two-level Electoral College, has been used many years in US presidential elections, and applied in areas such as image processing, pattern recognition, information retrieval, and other political elections. Our model will be the first in AI research that illustrates the stability of multi-level Electoral Colleges with respect to region/prefecture size and number of levels. Our theory will guide users of the applications about choosing an adequate number of levels, as well as an adequate region size for each level, when using a multilevel Electoral College to improve the performance of pattern recognition and information retrieval approaches. Developments made on the theoretical front will be applied in two important fields. (1) A face recognition surveillance system will be developed. This system will combine proven active infrared illumination technology for pupil location with a mathematical approach developed in this research program to match unknown faces to known faces on watch lists. The system is expected to be a significant improvement compared to any appearance-based face recognition system in recognition rate vs. false acceptance rate, and being disguise-proof. (2) An intelligent textual information retrieval system will be developed. This system will employ a novel approach, where a document is treated as a physical object which can be described from different "view angles", for document representation so that each document can be divided into blocks and blocks be further divided into subblocks. A vector space model will be employed as the basic strategy for matching a query with each lowest level "block" in a document, and an Electoral College will be employed in higher level blocks. The system is expected to have much better performance in precision vs. recall compared to other approaches. These systems can be further developed for use in national security systems and internet intelligent document retrieval systems respectively.

在理论方面，我们将研究多候选人选举中多级选举人团投票制度的稳定性。其最简单的形式是两级选举人团制，在美国总统选举中已经使用多年，并应用于图像处理、模式识别、信息检索和其他政治选举等领域。我们的模型将是人工智能研究中第一个说明相对于地区/县规模和级别数量的多层次选举学院稳定性的模型。我们的理论将指导应用程序的用户在使用多层选举团来提高模式识别和信息检索方法的性能时，如何选择足够数量的级别，以及每个级别的适当区域大小。在理论方面取得的进展将应用于两个重要领域。(1)开发人脸识别监控系统。该系统将结合经过验证的用于瞳孔定位的主动红外照明技术和本研究项目开发的数学方法，将观察名单上的未知面孔与已知面孔进行匹配。预计该系统在识别率、错误接受率、防伪装等方面，与任何基于外观的人脸识别系统相比，都有很大的提高。(2)开发智能文本信息检索系统。该系统将采用一种新颖的方法，将文档视为可以从不同“视角”描述的物理对象，用于文档表示，以便将每个文档划分为块，然后将块进一步划分为子块。将使用向量空间模型作为基本策略，将查询与文档中每个最低级别的“块”进行匹配，并在更高级别的块中使用选举人团。与其他方法相比，该系统有望在准确率和召回率方面有更好的表现。这些系统可以进一步发展，分别用于国家安全系统和互联网智能文件检索系统。