International Journal for Advanced Research in Science & Technology

The expanding modernity of Android malware has ended up a critical cybersecurity challenge, requiring progressed strategies for location and classification.Conventional machine learning (ML) strategies battle to viably identify advancing malware variations due to their complexity and the utilize of progressed pressing and muddling strategies.This paper presents an Computerized Android Malware Discovery utilizing Ideal Outfit Learning Approach (AAMD-OELAC), a novel strategy planned to upgrade the exactness and strength of Android malware location.The proposed approach coordinating three effective machine learning models: Slightest Square Back Vector Machine (LS-SVM), Bit Extraordinary Learning Machine (KELM), and Regularized Irregular Vector Utilitarian Connect Neural Arrange (RRVFLN), combined utilizing an outfit learning technique.This gathering approach leverages the qualities of each demonstrate to progress classification execution.Moreover, an Optimization calculation, Hunter-Prey Optimization (HPO), is utilized to fine-tune the hyper parameters of these models, guaranteeing ideal execution.The AAMD-OELAC strategy robotizes the whole malware location handle, from information preprocessing to classification, making it exceedingly productive for largescale arrangement.A comprehensive test assessment illustrates that the AAMD-OELAC strategy outflanks existing location strategies in terms of location exactness, exactness, review, and by and large discovery rate, especially for unused and complex malware variations.The comes about affirm that this gathering learning approach, improved by ideal parameter tuning, gives a strong arrangement to the advancing issue of Android malware location.The proposed framework can essentially make strides cyber security resistances, advertising an mechanized and exceedingly solid strategy for identifying noxious applications in genuine time.