3. ABSTRACT:
• Develop a robust framework for detecting malware in
Android applications using reverse engineering
techniques and machine learning algorithms.
• Android : The leading OS in smartphone technology
today.
• Our framework combines the power of machine
learning algorithms with reverse engineered
features to address this issue.
• By training the system on a large dataset of known
malware and benign applications.
• The results show a high detection rate (96.24%) with
a low false positive rate(0.3%).
4. EXISTING SYSTEM:
• Various methods proposed in related research to
improve malware detection in Android
applications.
• Some methods aim to increase accuracy, while
others focus on providing larger datasets or
utilizing different sets of features.
• The authors used the Random Forest algorithm
and introduced the PMDS approach for detecting
malware in Android applications.
• Their experiments demonstrated high accuracy
rates(94%) in detecting malware samples, and
PMDS proved effective in identifying previously
unknown malware with low false positive rates.
5. DISADVANTAGES:
• Considers requested permissions as
behavioral markers for detecting
potentially dangerous behavior in
unknown apps.
• The system is not implemented
Machine Learning Algorithm and
Ensemble Learning.
• The system is not implemented
Reverse Engineered Applications
characteristics.
6. PROPOSED
SYSTEM:
• we developed a new set of features
to detectAndroid malware, which
we tested on a large dataset.
• The results showed significantly
improved accuracy and a low false-
positive rate
• Our model is trained on the latest
malware samples, including the
most recent Android version.
7. SYSTEM REQUIREMENTS:
HARDWAREREQUIREMENTS:
• Processor : Pentium-IV.
• Ram : 4GB(min).
• Hard Disk : 20GB.
• Keyboard : Standard Windows Keyboard.
• Monitor : SVGA.
SOFTWAREREQUIREMENTS:
• Operating system : Windows 7 Ultimate.
• Coding Language : Python.
• Front-End : Python.
• Back-End : Django-ORM
• Designing : Html, CSS, JavaScript.
• Data Base : MySQL (WAMP Server).
8. CONCLUSION:
• Our framework presents an innovative approach to detecting malware in Android applications
using machine learning algorithms
• We offer an efficient solution aiming to safeguard user privacy and device integrity.