ANDROID MALWARE DETECTION MODEL USING STATIC AND DYNAMIC ANALYSIS WITH AI
International Research Journal of Modernization in Engineering Technology and Science · 2025
The wide and popular scope of the Android platform has been absorbed by both legitimate developers and official entities.This may reach a large number of security devices and other devices.Malware detection is one of the most important elements of operating systems and their security.Light from these diverse, fully developed and the most important point, we continue to explore the methods of detecting and combating Android malware detection and combating.Our proposed model for malware projects will detect and protect the Explorer system running on the crash system.Artificial intelligence and learning algorithms will be used for large facilities.High potential because malware is caused by many different factors.We aim to design and develop an effective malware agent that uses learning algorithms.The system will take advantage of static features and engineering techniques to improve malware recognition.Python programming language will be used for the proposed model.The system will work on the characteristics of static features and functional beauty engineering techniques and learning algorithms to improve malware recognition, and learn partners and complexity in rootkit data effectively, resulting in improved detection significantly.The project will provide an effective contributor researcher in identifying malware deletions, in addition to the Internet Protocol and its performance.In building the proposed system, we created three systems, the first of which uses machine learning, the second uses deep learning, and the third uses large language model (LLM) techniques, which was done by using the gpt-4 model.Through the three models, we seek to create malware detection systems that allow working within environments with different resources in terms of memory and processing.We seek to use the latest technologies in the field of artificial intelligence.As for using the LLM model, it seeks to evaluate the program, give reasons for rejecting the program, and provide an analysis of the reasons for program vulnerabilities.