Table 2 NF-UNSW-NB15 distribution.

From: A multilayer deep autoencoder approach for cross layer IoT attack detection using deep learning algorithms

Class

Count

Description

Benign

1550712

Normal harmless flows

Fuzzers

19463

An assault is when the hacker attempts to discover security holes by sending a flood of random data to crash the system.

Analysis

1995

A group of dangers that aim to compromise web applications using scripts, ports, and emails.

Backdoor

1782

A technique that may bypass security protections by responding to custom-built client applications

DoS

5051

Denial of service attacks aim to overwhelm a system’s resources, preventing users from accessing or using its data.

Exploits

24736

These are sets of instructions that utilize a known vulnerability to influence a host’s behavior.

Generic

5570

A technique that interferes with each block cipher and targets cryptography

Reconnaissance

12291

Also referred to as a probe, it is a method for learning more about a network host.

Shellcode

1365

Trojan horses are malicious programs that sneak into a system and hijack the host computer.

Worms

153

Attacks that infect further devices by self-replication