Phisherman

• 

  Phisherman.

• 

  Home Page UI

• 

  Suspicious email detected!

Inspiration

As avid computer science students, and users of technology, we've all faced some form of malware or suspicious emails. The inspiration for this project was rooted from a phishing email that one of our members received. It was difficult to recognize and was almost entirely a replica of a real email. Now, as someone without a tech background, it is likely that I would not be able to differentiate between a malicious and a real email. This sparked the idea to create an extension of chrome that allows users to simply check if their emails have suspicious content in them. As a result of using the Phisherman extension, the user gets familiar with suspicious emails which greatly protects their privacy.

What it does

The extension reads through all the emails present on the outlook or gmail website, converts the body of these emails into text files that are, when certain buttons are clicked, is sent to the backend server. The server functions as a verification tool that uses an AI model to analyze the email bodies and return a caution statement if there are suspicious emails present.

How we built it

Phisherman was developed using two major components, frontend and backend. The frontend design handled the extension drop down, logo, buttons and text, and event handlers. These functions were then linked to the python and java script backend that run the AI model.

Challenges we ran into

The biggest challenge we ran into was integrating the frontend with backend program. This was because the backend was using a REST API which made is more difficult to link it to the manifest.json file in the frontend. This was resolved by spending meticulous time tracking the flow of events and debugging our code.

Accomplishments that we're proud of

Our extension is compatible with both outlook and Gmail which are two of the most popularly used email websites and hence, it makes our product widely usable.

What we learned

We learnt what REST API is and how it can be integrated with basic codes and programs. Additionally, we understood the process behind building a full-stack web app, along with building and training an AI model.

What's next for Phisherman

There are several future enhancements such as email filter that only scans the email selected, allowing for user privacy. Another improvement would be marking the emails that are flagged as suspicious. More additions would be making a more dynamic extension that works with messaging apps and social platforms.

Built With

• css
• datasets
• flask
• html
• javascript
• joblib
• kaggle
• python
• rest
• sklearn