Improving spam email detection using deep recurrent neural network

<span>Nowadays the entire world depends on emails as a communication tool. Spammers try to exploit various vulnerabilities attack users with spam emails. While it is difficult prevent email attacks, many research studies have been developed in last decade an attempt detect These were conducted using machine learning techniques and types of neural networks. However, all their attempts highest accuracy acquired was 94.2% by random forest classifier. Deep demonstrated higher performance compared traditional algorithms. In this paper, deep recurrent network used determine whether email. After investigating different configurations for method, best setting that generated based Tanh activation function dropout rate equals 0.1 number epochs achieving 100. The proposed approach attained high 99.7% which surpassed (98.7%) obtained hybrid gated unit approach.</span>