SGER: Text Analysis and Pattern Detection: 3-D and Virtual Reality Environments

SGER：文本分析和模式检测：3D 和虚拟现实环境

• 批准号: 0741556
• 负责人: Lewis Lancaster
• 金额: $ 9.97万
• 依托单位: University of California-Berkeley
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2007
• 资助国家: 美国
• 起止时间: 2007-09-01 至 2009-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0741556&HistoricalAwards=false
• 关键词: SGER; Text; Analysis; Pattern; Detection

IIS - 0741556Lewis Lancaster, PIText Analysis and Pattern Detection: 3-D and Virtual Reality EnvironmentsAbstractThis project will explore the use of high dimensional visualization for analyzing text structure and patterns for scholars in the humanities. The use of computing in text study has not exploited virtual reality and 3D visualization to the degree that can significantly extend scholarly examination. This project suggests new, untried approaches to text analysis of significant foundational texts. The test document of study will be the Korean Buddhist Canon, the oldest complete set of the texts that make up the Buddhist canon for East Asia. The document consists of the more than 83,000 blocks of text which has been digitized with each characters marked up with a Unicode designation. Millions of glyphs will be converted into an image that will have the same metadata as the glyph rendering it capable of being linked to the appropriate description in the text. In this way patterns of visualized word distribution and frequencies can be used to examine more complex document structures. Ancient literature generally exhibits multiple structures that are critical to interpretation of origin and meaning. New methods for revealing and analyzing structure are of great interest to social scientists, linguists and humanists. These present primary challenges to computer science in multiple areas.

IIS -0741556刘易斯兰开斯特，PIText分析和模式检测：3-D和虚拟现实环境摘要本项目将探讨使用高维可视化分析文本结构和模式的学者在人文。 在文本研究中使用计算还没有利用虚拟现实和3D可视化的程度，可以显着扩展学术考试。 这个项目提出了新的，未经尝试的方法来文本分析的重要基础文本。 研究的测试文件将是韩国佛教经典，这是构成东亚佛教经典的最古老的完整文本。 该文档由超过83，000个文本块组成，这些文本块已数字化，每个字符都标记有Unicode名称。数以百万计的字形将被转换成一个图像，该图像将具有与文本相同的元数据，使其能够链接到文本中的适当描述。 通过这种方式，可以使用可视化的单词分布和频率模式来检查更复杂的文档结构。 古代文学通常表现出多重结构，这对解释起源和意义至关重要。 揭示和分析结构的新方法引起了社会科学家、语言学家和人文学家的极大兴趣。 这些在多个领域对计算机科学提出了主要挑战。