TABLE OF CONTENTS Previous chapters discuss the main classes of compression methods: RLE, statistical methods, and dictionary-based methods. There are data compression methods that are not easy to classify and do not clearly belong in any of the classes discussed so far. A few such methods are described here.
The Burrows-Wheeler method (Section 8.1) starts with a string S of n symbols and scrambles (i.e., permutes) them into another string L that satisfies two conditions: (1) Any area of L will tend to have a concentration of just a few symbols. (2) It is possible to reconstruct the original string S from L.
The technique of symbol ranking (Section 8.2) uses context to rank symbols rather than assign them probabilities.
ACB is a new method, based on an associative dictionary (Section 8.3). It has features that relate it to the traditional dictionary-based methods as well as to the symbol ranking method.
Section 8.4 is a description of the sort-based context similarity method. This method uses the context of a symbol in a way reminiscent of ACB. It also assigns ranks to symbols, and this feature relates it to the Burrows-Wheeler method and also to symbol ranking.
The special case of sparse binary strings is discussed in Section 8.5. Such strings can be compressed very efficiently due to the large number of consecutive zeros they contain.
Compression methods that are based on words rather than individual symbols are the subject of Section 8.6.
Textual image compression is the topic of...
