Modelo de minimização de entropia da informação aplicado à compressão de dados digitais

Abstract

In this work, the results of the experimental evaluation of a proposed model of information entropy minimization in digital data compression processes are presented. During the research, mathematical models and alphabets used in encoding digital data were analyzed. The approach adopted was the encoding of digital data in a geometric structure totally symmetric in space. During the execution of the experiments, several aspects related to the data encoding process were monitored, such as computational effort, compression rate, initial and final entropy. As a result of this work, the effectiveness of the data entropy reduction model was verified, which reduces the uncertainty of information in digital data clusters to zero. In this way, the encoding processes resulting from this model will output a constant amount of data, regardless of the size of the source sets and allow lossless decoding of data, thus reducing the entropy of information to levels close to zero.