Centurian Fraud Detector

Inspiration

The inspiration behind creating this solution stemmed from a recognition of the inefficiency and impracticality of manually reviewing every transaction. We realized the tremendous potential to streamline and enhance the auditing process by leveraging the power of machine learning and Microsoft Azure.

What it does

This web application uses machine learning and Azure to find fraudulent transactions within a given list.

How we built it

We started by training and creating a machine learning module using decision forests using Tensorflow. We deployed the back end and front using Azure Blob and Azure functions, we then deployed our module using the Azure machine learning studio.

Challenges we ran into

We had not used Azure before, so deploying and configuring this was certainly a challenge. We also had no experience making a custom ML module. Having to learn these in a short amount of time required great effort.

Accomplishments that we're proud of

Learning about machine learning and decision trees, Azure and DevOps are a few of the things we are proud of.

What we learned

Machine learning, Azure and DevOps, along with teamwork and time management.

What's next for Centurian Fraud Detector

Creating a database for users to permanently have access to their records.

Built With

• azure
• css
• figma
• html
• javascript
• keras
• python
• tensorflow