Abstract:
This paper presents an implementation of a method for text compression, first described by Shannon in 1951 and later, in 1997, by Fenwick. Unlike other compressors, which exploit symbol frequencies in order to assign shorter codes to more frequent symbols, this technique prepares a list of probable symbols to follow the current one, ordered from most likely to least likely. Tests were conducted on various input sources and the results are shown here. The implementation supports further optimizations, as it will be explained.