Abstract
The ever-increasing growth of online services and smart connectivity of devices have
posed the threat of Android Malware to Mobile app system, android-based smart
phones, Internet of Things(IoT)-based systems. The anti-Android Malware software plays
an important role in order to safeguard the system resources, data and information
against these Android Malware attacks. Nowadays, Android Malware writers used
advanced techniques like obfuscation, packing, encoding and encryption to hide
the malicious activities. Because of these advanced techniques of Android Malware
evasion, traditional Android Malware detection system unable to detect new variants of
Android Malware. Cyber security has attracted many researchers in the past for
designing of Machine Learning (ML) or Deep Learning (DL) based Android Malware
detection models. In this study, we present a comprehensive review of the literature on
Android Malware detection approaches. The overall literature of the Android Malware
detection is grouped into three categories such as review of feature selection (FS)
techniques proposed for Android Malware detection, review of ML-based techniques
proposed for Android Malware detection and review of DL-based techniques proposed
for Android Malware detection. Based on literature review, we have identified the
shortcoming and research gaps along with some future directives to design of an
efficient Android Malware detection and identification framework.
Android Malware
Android Malware is any software intentionally designed to cause damage to a Mobile
app, server, client, or Mobile app network. A wide variety of Android Malware types exist,
including Mobile app viruses, worms, Trojan horses, ransomware, spyware, adware,
rogue software, wiper and scareware.
Types of Android Malware : Trojan horse, Virus, Adware, bots, bugs, rootkits, spyware
Android Malware DETECTOR :
Android Malware detection is the process of scanning the Mobile app and files to
detect Android Malware. It is effective at detecting Android Malware because it involves
multiple tools and approaches. It's not a one way process, it's actually quite complex.
Android Malware Detection
Android Malware detection is the process of scanning the Mobile app and files
to detect Android Malware. It is effective at detecting Android Malware because
it involves multiple tools and approaches. It's not a one way process, it's
actually quite complex.
Android Malware Attacks and How to
Prevent Them
1. Viruses
• Viruses require human intervention to propagate.
• Once users download the malicious code onto their devices -- often delivered via malicious
advertisements or phishing emails the virus spreads throughout their systems.
• Viruses can modify Mobile app functions and applications; copy, delete and exfiltrate data.
2. Adware:
• It is capable of downloading or displaying advertisements to the device user.
• Not steel any data from the system but it forcing users to see ads.
• Some Irritating forms of adware display browser pop-ups that cannot be closed.
3. Ransomware
• Ransomware locks or encrypts files or devices and forces victims to pay a ransom in exchange
for reentry. While ransomware and Android Malware are often used synonymously, ransomware is
a specific form of Android Malware.
• Types of Ransomware : Locker ransomware, Crypto ransomware, Triple extortion
ransomware,
4. Rootkits
• A rootkits is malicious software that enables threat actors to remotely access
and control a device.
• Rootkits facilitate the spread of other types of Android Malware, including
ransomware, viruses and keyloggers.
• Rootkits often go undetected, because once inside a device, they can deactivate
antiAndroid Malware and antivirus software.
• Rootkits typically enter devices and systems through phishing emails and
malicious attachments.
5. Spyware
• Spyware is Android Malware that downloads onto a device without the user's
knowledge.
• It steals users’ data to sell to advertisers and external users.
• Spyware can track credentials and obtain bank details and other sensitive data.
• It infects devices through malicious apps, links, websites and email attachments.
How to prevent Android Malware attacks
Strong Cyber hygiene is the best defense against Android Malware attacks. The
premise of cyber hygiene is similar to that of personal hygiene: If an organization
maintains a high level of health (security), it avoids getting sick (attacked).
Cyber hygiene practices that prevent Android Malware attacks include the following:
• Follow email security best practices.
• Deploy email security gateways.
• Avoid clicking links and downloading attachments.
• Implement strong access control.
• Require multifactor authentication.
• Use the principle of least privilege.
• Adopt a zero-trust security strategy.
• Monitor for abnormal or suspicious activity.
Android Malware Symptoms
Mobile apps, they all can produce similar symptoms. Mobile apps that are
infected
with Android Malware can exhibit any of the following symptoms:
• Increased OS usage
• Slow Mobile app or mobile app speeds
• Problems connecting to networks
• Freezing or crashing
• Modified or deleted files
• Appearance of strange files, programs, icons
• Programs running, turning off, or reconfiguring themselves (Android
Malware will often reconfigure or turn off antivirus and firewall programs)
• Strange Mobile app behavior
• Emails/messages being sent automatically and without user's knowledge
(a friend receives a strange email from you that you did not send)
Existing Systems for Android Malware
detection using machine learning techniques
• Implement Machine Learning Pipeline: Leverage a machine learning pipeline for
Android Malware detection, as illustrated in the provided figure, to enhance the
system's capabilities.
• Utilize Advanced Algorithms: Apply advanced machine learning algorithms to
analyze large volumes of data effectively, enhancing Android Malware detection
accuracy.
• Incorporate Dynamic Android Malware Detection: Focus on dynamic Android
Malware detection to adapt to evolving threats, considering the progressive changes in
Android Malware behavior.
• Explore Automated System-Level Detection: Investigate automated system-level
Android Malware detection, exploring fundamentals and the current status quo in
machine learning-based detection systems.
• Consider Proposed Techniques: Evaluate proposed methods, like the one
demonstrating effectiveness in Android devices for automated Android Malware
detection.
• Regularly Update Models: Keep machine learning models updated to stay resilient
against emerging Android Malware threats.
EXISTING SYSTEMS
Android Malware detection by using window api sequence and
machine learning
Detecting unknown malicious code by applying classification
techniques on oppose patterns
Detecting scareware by mining variable length instructions sequence
Accurate adware detection using oppose sequence extraction
Detection of spyware by mining executable files
Detection by using neural networks on the Android Malware
MACHINE LEARNING
Machine learning is a method of data analysis that automates analytical model
building. It is a branch of artificial intelligence based on the idea that systems
can learn from data, identify patterns and make decisions with minimal human
intervention.
Types of machine learning
Supervised learning
Unsupervised learning
Reinforcement learning
PROPOSED SOLUTION WITH
ALGORITHMS
Machine learning can easily identify the Android Malware in the data
and datasets
Different types of machine learning algorithms are applied such as :
DECISION TREE
SVM
Random forest
XG boost
CONCLUSION
A Android Malware is critical threat to user Mobile app system in
terms of stealing confidential information or disabling security.
This project present some of the existing machine learning algorithms
directly applied on the data or datasets of Android Malware
It explains the how the algorithms will play a role in detecting Android
Malware wit high accuracy and predictions
We are also using data science and data mining techniques to
overcome the drawbacks of existing system
REFERENCES
https://en.wikipedia.org/wiki/Android Malware
https://en.wikipedia.org/wiki/Machine_learning
https://en.wikipedia.org/wiki/Supervised_learning
https://en.wikipedia.org/wiki/Spamming
https://
www.researchgate.net/publication/343499527_Project_report_Android
Malware_analysis
https://
towardsdatascience.com/Android Malware-detection-using-deep-learn
ing-6c95dd235432
Thank You
