Abstract
In this paper, Data compression plays a significant role and is necessary to minimize the storage size and accelerate the data transmission by the communication channel object, The quality of dictionary-based text compression is a new approach, in this concept, the hybrid dictionary compression (HDC) is used in one compression system, instead of several compression systems of the characters, syllables or words, HDC is a new technique and has proven itself especially on the text files. The compression effectiveness is affected by the quality of hybrid dictionary of characters, syllables, and words. The dictionary is created with a forward analysis of text file a. In this research, the genetic algorithm (GAs) will be used as the search engine for obtaining this dictionary and data compression, therefore, GA is stochastic search engine algorithms related to the natural selection and mechanisms of genetics, the s research aims to combine Huffman's multiple trees and GAs in one compression system. The Huffman algorithm coding is a tactical compression method of creating the code lengths of variable-length prefix code, and the genetic algorithm is algorithms are used for searching and finding the best Huffman tree [1], that gives the best compression ratio. GAs can be used as search algorithms for both Huffman method of different codes in the text characters leads to a difference in the trees, The GAs work in the same way to create the population, and can be used to guide the research to better solutions, after applying the test survive and their genetic information is used.
The proposal is a new multi-trees approach to compute and present the best Huffman code trees of characters, syllables, and words to get the best possible of minimization compression [2]. The frequency increases to the text have an impact on compression rate. This approach generates more distinctive code words and efficient of space consumption and computation for compression rate without losing any data on the original files, the experiments and results show that a significant improvement and accuracy can be achieved by using the code words derived [3].