Abstract:
Malware is any programme that is intended to cause harm to other software or
hardware. The growth of malware poses a grave threat to cyber security. Malware
can result in sluggish computation, frequent system crashes, memory consumption,
etc. Protection should increase at the same rate as technological development. In the
field of cyber security, machine learning has played a significant role in the detection of
malware. In this proposed system, machine learning algorithms such as random forest
and gradient boosting are utilised to combat the rising prevalence of malware. Random
forest creates a collection of decision trees, each of which is a fundamental forecast,
and combines the results into a single output. In gradient boosting, successive decision
trees are constructed to address the deficiencies of the preceding trees. Gradient
boosting accumulates results along a route. Using datasets of malware and non-
malware samples, a malware classifier capable of detecting the presence or absence of
malware is constructed, and the accuracy of both algorithms is evaluated to determine
which method is more efficient.
