I am planning to build a successful voting system build using blockchains that is secure, transparent and stable. The novely in my approach is that I want to make use of AI to find trends in the collected data, as the polls go on, and help bring insight into knowledge such as, predicted voter turnout, predicted winner, demographic analysis etc.
I first plan on scaffolding a basic voting system using the tools I am provided with on Azure. The next step will be to ensure that it is secure, and has more functionality to make it more flexible. Next I want to make this solution as scalable as possible. Regarding the use of AI, I want to train a model(mostly a DNN) with previous election data, and use this trained model to bring insight into voting trends for an election that occurs in realtime. For this, the Data field in each node of the Blockhain must be elaborate with information that is also not too revealing.
- Build a election system
- Make it robust and compatible with current voting schemes
- Train a DNN with election data
- Merge DNN with blockchain flow
- See what kinds of features can be captured from data field of DNN in realtime
- Try and generate visualizations of realtime trends.
The exact dataset to be used is undecided
- Azure Servies
- Python Tensorflow DNN
- Possible Node Server used for testing purposes.