Skip to content

Creating a LSTM model through assembly level smali code of Apk Files, after rigourrous preprocessing and code Analysis.

Notifications You must be signed in to change notification settings

Tejas3772/Research-Malware-LSTM

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Research-Malware-LSTM

Creating a LSTM model through assembly level smali code of Apk Files, after rigourrous preprocessing and code Analysis.

LSTM-Based Behavior Pattern Generation:

In this crucial phase of our project, we leverage Long Short-Term Memory (LSTM), a type of recurrent neural network (RNN), to generate behavior patterns from the analyzed data. The LSTM model is a powerful tool for understanding sequential data and capturing temporal dependencies, making it particularly well-suited for our task.

  1. Sequence Understanding: The LSTM model takes as input the preprocessed and structured data, which represents the behavior of mobile applications over time. It excels in learning from sequences, making it possible to capture the dynamic nature of app actions.

  2. Temporal Dependencies: LSTM's unique ability to remember and learn from past observations allows it to identify patterns and relationships in the data that might not be apparent through traditional methods.

  3. Pattern Generation: Through training on our dataset, the LSTM model learns to generate behavior patterns that are indicative of both genuine and fake apps. These patterns provide valuable insights into the typical actions and interactions of mobile applications.

  4. Enhanced Security Insights: By analyzing the behavior patterns generated by LSTM, we gain deeper insights into the characteristics of genuine and malicious apps. This information is critical for classification and forms the basis for our security measures.

  5. Continuous Learning: As part of the future scope, we aim to further enhance the LSTM model's ability to adapt and learn from new data continuously. This ensures that our security system remains up-to-date and effective in identifying emerging threats.

Overall, the LSTM-based behavior pattern generation is at the core of our project's success, enabling us to understand, classify, and ultimately enhance the security of mobile applications based on their behavior over time.

About

Creating a LSTM model through assembly level smali code of Apk Files, after rigourrous preprocessing and code Analysis.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published