Internet of Things Assisted Automated Ransomware Recognition using Harmony Search Algorithm with Deep Learning

The Internet of Things (IoT) denotes a networking system of interrelated equipment's that is capable of gathering and interchanging information in the Internet system with no human interference. A few such devices are sensors, wearables, smart appliances, industrial machines, cameras, and much more. Under the cyber safety context, IoT gadgets can be employed for gathering useful information and giving added data for recognizing possible ransomware challenges. Ransomware is a kind of malevolent software which will be encrypting the data of a victim, making it unavailable until the attacker is paid off. Automatic detection of ransomware encompasses the utilization of Artificial Intelligence (AI), Machine Learning (ML), or other automatic techniques for classifying and detecting ransomware outbreaks depending on behaviours, particular patterns, or characteristics. This manuscript presents an IoT-assisted Automated Ransomware Recognition using a Harmony Search Algorithm with a Deep Learning (ARR-HSADL) approach. The ARR-HSADL technique presents three major phases of operations and intends a proper ransomware classification. At the initial stage, the ARR-HSADL technique uses HSA to elect a set of features. Next, the ARR-HSADL technique utilizes the Long Short-Term Memory (LSTM) approach for the ransomware detection process. Finally, the ARR-HSADL technique uses a Root Mean Square Propagation (RMS Prop) optimizer for the procedure of tuning. The investigational output of the ARR-HSADL methodology is experimented on the ransomware dataset. The obtained values inferred that the ARR-HSADL methodology exhibits its betterment with compared models.