Mining high-dimensional and graph data using spectral analysis

• Main Author: Li, Wenyuan
• Other Authors: Ng Wee Keong
• Format: Theses and Dissertations
• Published: 2008
• Subjects: DRNTU::Engineering::Computer science and engineering::Information systems::Information storage and retrieval
• Online Access: https://hdl.handle.net/10356/2360
• Institution: Nanyang Technological University

Although the field of data mining has seen major advancements in the past fifteen years, algorithms for handling complex data (with high dimensionality or complex graph structures) are only becoming the mainstream in recent years. To address the difficulties of mining complex data, we argue that a right understanding of data characteristics (i.e., the general information of the data that is not particularly designed for any specific data mining task, but might enhance many types of data mining tasks) is important. The objective of this thesis is to study and exploit spectral information to provide quick insights into how data characteristics are beneficial to specific applications. We study issues concerning the design of how spectral information can be integrated into the needs of different types of analysis.