Android App Analyzer

Analyzes android application files (APKs), extract in-depth information and checks through AI Models, libraries, and tools to see if app is suspicious, harmful or good, and gives the app a score.

Comment 1

Inspiration

Lack of details while downloading and using apps are raising concerns about smartphone users' privacy. There is a need of a comprehensive tool to analyze applications that checks through previously collected data, external malware analysis tools, and with utilization of Artificial Intelligence to marks apps as suspicious or good.

What it does

The app authenticates user, provides a dashboard to view old apk files uploaded and their meta-data and permissions requirement. It offers a feature to upload an apk file, and returns metadata and permissions required of apps, this helps in giving knowledge to users about the app. It then saves the apk and extracted data in the history for that particular user and shows it on the dashboard. (Note: The app is not yet fully built, integrations with powerful tools like VirusTotal through its API is being done at the moment, AI model is being trained and developed on the android apps data, and other functions are being built or integrated at the moment, that will make this app into an all-in-one comprehesive and powerful tool to analyze apps and spot all kind of suspicious things in them including malware, viruses, spyware etc.)

How we built it

It was built using flask (Python) and its libraries at the backend, general JavaScript at the frontend, HTML and CSS for the structure and styling of the webapp, while SQLite 3.0 as the database. The AI model is being trained and developed with Pandas, Numpy among other libraries.

Challenges we ran into

Routing was difficult for me being a beginner, however I overcame the challenge and learned through practice the routing concepts. Integrations with third-party tools are another challenge, which I am working on at the moment. Deployment to Azure and CI/CD implementation, however I learned that quickly.

Accomplishments that we're proud of

Use of AI, Third-party tools, data-driven solution to acheive a comprehensive system to analyze apps and keep smartphones secure.

What we learned

Working in Flask with databases. Integration with third-party tools using APIs. Integration of AI with webapp. CI/CD and deployment in cloud (Azure).

What's next for Android App Analyzer

Complete integration of AI model, third-party tools, and more enhanced UI.

Share this project:

Updates

posted an update —

Currently working on following additions for this project:

  • Developing an AI model on extensive data to classify malicious and safe apps, and then tell users.
  • Utilizing third-party services for more in-depth analysis of APKs.
  • Developing Android App to monitor real-time usage of apps, and track their workings and analyze if the apps are working in a suspicious way.

Log in or sign up for Devpost to join the conversation.