Malware Detection Based on the Feature Selection of a Correlation Information Decision Matrix

Smartphone apps are closely integrated with our daily lives, and mobile malware has brought about serious security issues. However, the features used in existing traffic-based detection techniques have a large amount of redundancy useless information, wasting computational resources training models. To overcome this drawback, we propose feature selection method; core method involves choosing selected based on high irrelevance, thereby removing redundant features. Furthermore, artificial intelligence implemented achieved outstanding ability. almost all models deep learning include pooling operations, which lead to loss some local information affect robustness model. We also designing model for malicious traffic identification capsule network. The main difference between network neural is that neuron outputs scalar, while vector. It more conducive saving information. verify effectiveness method, it from three aspects. First, use four popular machine algorithms prove proposed method. Second, compare convolutional superiority Finally, another state-of-the-art technique; accuracy recall increased by 9.71% 20.18%, respectively.