Cyber Forensic Investigation in IoT Using Deep Learning Based Feature Fusion in Big Data.

Cyber Forensic Investigation in IoT Using Deep Learning Based Feature Fusion in Big Data.

BIG data

Cybercrime detection

Deep belief network

DEEP learning

Deep quantum neural network

FORENSIC sciences

Internet of things

INTERNET of things

MAPREDUCE

MULTISENSOR data fusion

OPTIMIZATION algorithms

The Internet of Things (IoT) device is becoming universal domain and its success cannot be ignored, but its threats on IoT devices increases concurrently. The Cyber-attacks are becoming the component of IoT affecting user's life. The professionals are forced to sift huge data to unveil and manage litigations. Hence, secure IoT is required for comprehending attacks. A model is presented for discovering cyber attack considering feature fusion. The routing of data towards Base Station (BS) is done with the Fractional gravitational search algorithm (FGSA). At BS, cybercrime detection is done, wherein data is splitted with enhanced Fuzzy c-means clustering (FCM) considering the MapReduce model. In mapper, the feature fusion is done with mutual information and the Deep Quantum Neural Network (DQNN), while reducer performs cybercrime detection. The Fractional Mayfly Shepherd Optimization (FrMSO)-based Deep Belief Network (DBN) is devised for describing the digital examination to notice and trace behaviors of attacks in IoT. Here, the training of DBN is done by the proposed FrMSO algorithm, which is developed by integrating the Fractional Calculus (FC), Mayfly Optimization Algorithm (MA), and the Shuffled shepherd optimization Algorithm (SSOA). The developed model helps to employ the weights of DBN with FrMSO for determining and tracing the abnormal aspects in IoT. The FrMSO-based DBN presented elevated precision of 96.4%, recall of 98.3% and F-measure of 95.4% respectively. [ABSTRACT FROM AUTHOR]

16-29

1

30

Thapaliya, Suman

Sharma, Pawan Kumar

International Journal of Wireless Information Networks

2023

journalArticle

10689605

https://libproxy.lamar.edu/login?url=https://search.ebscohost.com/login.aspx?direct=true&

db=syh&

AN=163149583&

site=ehost-live