Thesis History And Details
From 1984 to 1989 I was a Ph.D. candidate in the Department Of Computer Science at the University of Adelaide. My supervisor was Bill Beaumont. Tao Li acted as temporary supervisor for about a year. I submitted my dissertation on 30 June 1989. My thesis examiners were Glen Langdon and Moshe Zukerman. The thesis was accepted on 16 November 1989 and I was awarded the degree on 30 April 1990. The number of the thesis at the Barr-Smith Library at the University of Adelaide is BSL-09PH-W7262.
In 1990/1991 my thesis was published as a book by Kluwer Academic Publishers. The book subsequently sold out (in 1992) and was reprinted (as reported in Lumen "Ph.D. Thesis Sold Out", Lumen 21(5) p.6, 8 May 1992, (The University of Adelaide News Magazine))
One of the reasons I think that the book sold well was because in addition to being a Ph.D. thesis bringing new ideas and research results, it was also a solid reference work. The book contains a 100 page overview of the field of text data compression which still remains a good introduction to the field of text compression.
In addition, the thesis is rich in experimental results and includes dozens of graphs of the performance of Markov text compression algorithms running under a wide range of conditions. It is an essential reference for anyone designing a text compression algorithm of any kind.
Here is an abbreviated table of contents:
Chapter 1: Introductory Survey................. 1 Chapter 2: The DHPC Algorithm.................. 107 Chapter 3: A Classification of Adaptivity...... 125 Chapter 4: An Experimental Adaptive Algorithm.. 145 Chapter 5: A Multimodal Algorithm.............. 245 Chapter 6: Applications to User Interfaces..... 283 Chapter 7: Conclusions......................... 305
Chapter 1 gives a comprehensive overview of the field of text data compression. Chapter 2 describes a simple Markov algorithm called DHPC. Chapter 3 investigates the use of adaptivity in data compression algorithms. Chapter 4 describes the SAKDC algorithm and also contains a wealth of experimental data. Chapter 5 introduces a new class of adaptivity in data compression algorithms (multimodal adaptivity) and describes the MMDC algorithm which implements it. Chapter 6 shows how data compression techniques can be applied in user interfaces.
ADAPTIVE DATA COMPRESSION reviews the field of text data compression and then addresses the problem of compressing rapidly changing data streams. To compress such streams, the compressor must continuously revise its model of the data, and sometimes even discard its model in favour of a new one.
Copyright © Ross N. Williams 1996-1997. All rights reserved.