In actuarial practice, actuaries need to sift through many claims for insurance to detect fraud. Although there are data analysis techniques for this, actuaries cannot go through all the claims and usually rely on a sample to determine the number of claims that might be fraudulent. Some fraudulent claims still slip through the cracks even with the techniques that the industry has right now. We found that blockchain technology could help with this since the blocks are immutable and can be verified by sources other than the policyholder themselves. We decided to work with the auto insurance industry first.
What it does
Puts claim requests and past history of the services used for the car on a blockchain. Data analysis techniques can be used on the blocks to detect suspicious claims.
How we built it
Blockchain was built with node.js. The website was built with HTML5, CSS, JS, Sass.
Challenges we ran into
We had a lot of bugs with the blockchain.
Accomplishments that we're proud of
We have learned how to code a blockchain in two days.
What we learned
Veronica learned how to do front-end web development. Maggie learned how to code a blockchain, along with improving her node.js command.
What's next for AutoBlock
To integrate machine learning onto the blockchain.