Text Compression Algorithm Research Articles

Text compression as rule based pattern recognition

A novel algorithm for text compression is developed which consists of two parts. In the first part the text sample is encoded by the Ziv-Lempel asymptotic structured codebook. In the second part decoding is carried out by a novel rule based (RB) algorithm which improves the compression ratio for a given text sample by up to 50% when compared to the full text search as required by the Ziv-Lempel decoding algorithm.

A systematic approach to compressing a full-text retrieval system

This article reports on a variety of compression algorithms developed in the context of a project to put all the data files for a full-text retrieval system on CD-ROM. In the context of inexpensive pre-processing, a text-compression algorithm is presented that is based on Markov-modeled Huffman coding on an extended alphabet. Data structures are examined for facilitating random access into the compressed text. In addition, new algorithms are presented for compression of word indices, both the dictionaries (word lists) and the text pointers (concordances). The ARTFL database is used as a test case throughout the article.

A text compression algorithm

We present an algorithm using statistical information about word and digraph frequencies as well as the structural pattern concerning words with a common root and different suffixes. The single or composed most frequent words, included in a suitable dictionary, are encoded with a text pointer; the remaining words, where possible, are encoded according to the most frequent digraphs included in a different vocabulary

CAI with a restricted vocabulary

Text compression techniques can be applied to computer assisted instruction (CAI) text to reduce both computer storage requirements and volume of transmission between the computer and student terminals. One of the more efficient text compression algorithms requires a large dictionary. containing all the words in the text. Unfortunately, the use of a large dictionary is not feasible with many current CAI systems. This study investigates some implications of authoring CAI text using words from a short dictionary established prior to writing the text. A dictionary of approximately one thousand words was defined. Author subjects wrote two sets of CAI text: one in natural English and one using words from the short dictionary. The natural English text was compressible to about two-thirds of the normal computer storage requirements; the restricted English, to less than one-third. There was no significant difference in the instructional value of the different types of text.