New text compression method based on direct solution method neural networks

Abstract EN

The aim of this work is to create a text compression technique using a single layer neural networks trained using direct solution method.

In this work, the text is made as desired output and the weights are pre-defined then, by using the direct solution method, we calculate the input vector.

This process, exhibits decreasing the number of the input neurons, the computing size and memory space.

The present work has been demonstrated through several experiments and show very promising results in compression rates as well as in the reconstructed text.