文档介绍：High Performance Data Mining in Time Series:
Techniques and Case Studies
by
Yunyue Zhu
A dissertation submitted in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Department puter Science
New York University
January 2004
Dennis Shasha
°c Yunyue Zhu
All Rights Reserved, 2004
To my parents and Amy, for many wonderful things in life.
iii
Acknowledgments
This dissertation would never have materialized without the contribution of
many individuals to whom I have the pleasure of expressing my appreciation
and gratitude.
First of all, I gratefully acknowledge the persistent support and encourage-
ment from my advisor, Professor Dennis Shasha. He provided constant aca-
demic guidance and inspired many of the ideas presented here. Dennis is a
superb teacher and a great friend.
I wish to express my deep gratitude to Professor Ernest Davis and Profes-
sor Chee Yap for serving on my proposal and mittees. Their
comments on this thesis are precious. I also thank the other members of my
mittee, Professor Richard Cole, Dr. Flip Korn and Professor
Arthur Goldberg, for their interest in this dissertation and for their feedback.
Rich interactions with colleagues improve research and make it enjoyable.
Professor Allen Mincer has both introduced me to high-energy physics and
arranged the access to Milagro data and software. Stuart Lewis has helped with
many exciting ideas and promising introductions to the ic Resonance
munity. Within the database group, Tony Corso, Hsiao-Lan Hsu,
Alberto Lerner, Nicolas Levi, David Rothman, David Tanzer, Aris Tsirigos,
Zhihua Wang, Xiaojian Zhao have lent both voices and helpful suggestions in
iv
the course of this work. This is certainly not plete list. I am thankful for
many friends with whom I share more than just an academic relationship.
Rosemary Amico, Anina Karmen and Maria L. Petagna performed the ad-
ministrative work required for this research. They were vital