Scribd
Upload
21 views3 pages

Android Malware Detection Using Deep Learning

The document proposes using a convolutional neural network with deep learning to classify Android applications as malware or non-malware. It discusses the objectives, disadvantages of existing systems, the proposed system using deep learning, and the advantages and software requirements.

Uploaded by

NoviTech
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
21 views3 pages

Android Malware Detection Using Deep Learning

The document proposes using a convolutional neural network with deep learning to classify Android applications as malware or non-malware. It discusses the objectives, disadvantages of existing systems, the proposed system using deep learning, and the advantages and software requirements.

Uploaded by

NoviTech
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

Android Malware Detection using Deep Learning

Objective:

Malware is one of the major issues regarding the operating system or in the
software world. The android system is also going through the same problems.The
main objective of this project is to create a model based on deep learning that can
automatically identify whether an android application is malware infected or not
without installation.

Existing System:

Traditionally Numerous malware detection tools have been developed,


but some tools are may not able to detect newly created malware application and
unknown malware application infected by various Trojan, worns, spyware
Detecting of large number of malicious application over millions of android
application is still a challenging task using traditional way. In existing, Non
machine learning way of detecting the malicious application based on
characteristics, properties, behavioral.

Disadvantages:

 Identification of newly updated or created malicious application is hard to


find out.

 Non Machine learning approaches are not reliable and efficient

 In Existing approaches covers only 30 permissions out of 300 app


permissions, due to this limited apps permissions different types of attacks
can occurs.
Proposed System:

In this project,the proposed system is based on Convolutional Neural Network


with deep learning for classification.This algorithm inspired by biological neural
networks (where the brain is considered particularly important in the central
nervous system) and are used in statistics and cognitive science. Theseare
represented by the interconnection of neural systems from various input
variables to the output, and it can be represented as mathematical functions that
are configured to represent complex relationships between inputs (independent
variables) and outputs (dependent variables).

Block Diagram:

Data Data
Dataset
Preprocessing Spliting

Train Data Test Data

Data Training
process

Output Model
Creation

Test Data
Advantages:

 Improves the percentages of detection malicious application.


 Deep learning is better efficient than Non machine learning algorithm.
 Able to detect new malware android applications.
 We only need to consider 22 out of409 permissions to improve the runtime
performance upto 91 %.

Software Requirements:-

 Windows 7,8,10
 Python(Version-Above 3.9)
 Ram(4GB or 8GB)
 Processor(i3 or i5)

You might also like