Safe Cash

Inspiration

Our inspiration is from PNP Coin, PNP Coin provides us with regulated cryptocurrency support. But it doesn't promote other cryptocurrencies to be regulated, which results in centralized coin in some countries.

What it does

Our platform allows users to still make transactions in a decentralized way but also provides a fraud detection system, ensuring that no fraudulent transaction is performed illegally. Our machine learning algorithm predicts the fraudulent transaction based on previous fraudulent transactions.

How we built it

We made a website using React.js and Tailwind CSS. We wrote a smart contract using Solidity. The website asks for To address ID and Amount. This is further sent to our fraud-detecting algorithm and it predicts and responds back to the website whether its fraudulent transactions are not, Based on its response the website decides whether or not transactions should continue. This allows all cryptocurrencies to make transactions in a regulated manner.

Challenges we ran into

As Blockchain is a new technology, we are lacking materials to study and develop the same applies to machine learning, Our first challenge was to create our dataset and test all the machine learning algorithms on the dataset and as there are multiple algorithms available we are confused to select which algorithm that could be used for our project,

Accomplishments that we're proud of

As of now, we have created a website that can make transactions seamlessly in the Ethereum blockchain network , We tested the website using Hardhat. We have implemented the logistic regression in Machine learning with R to detect fraudulent transactions.

What we learned

We learned website development using React.js and to write smart contracts using solidity . And using R we learned how to work on data frames and apply logistic regression on data frames to predict the fraudulent transaction

What's next for Safe Cash

Following up, we are going to use distributed neural networks which comes under deep learning to predict fraudulent transaction. Next, we are going to test our platform in the main network to find and fix the bugs and glitches. Improve UI/UX to make an immersive user experience.