In this work, a study on the use of a type of genetic algorithm for Android operating purposes. This application involves training the algorithm first on a sample set of samples from safe, fashionable, and samples from malicious applications, so that the developed model can then detect the type of application in the testing phase. Two scales were evaluated mainly with a score of 95.89%, and the detection rate with a score of 99.11%, both scales accuracy and F1 score were used in the evaluation as well. The model study has proven that the model of algorithms is characterized by low complexity and high speed of implementation. And also the high image has been tested in detection. The dissertation has been successfully accepted.