Coding the Tax Code: Regulations to Formalism
The STEALTH Project aims to model and represent regulations in the IRC simulate transactions and apply AI techniques to optimize tax output and risk of audit. Thus the result is a system that can predict new tax evasion strategies that might arise with changes in regulation. The first component of the system requires a formal representation of regulations. It would seem impractical to directly read and convert a large number of regulations (human-readable text) to machine-readable code that can then be used for simulation. Hence the main focus of this project is to develop an automatic parsing system for translating these regulations into a formalism. Our strategy is to experiment with NLP algorithms develop such a system to perform this parsing and translation.

I am participating in the SuperUROP program because I want to take part in exciting research and explore innovative techniques in the field of Natural Language Processing. Although I do not have much experience in this field I hope to learn as much as possible before and during the project; I will also be taking 6.806 in the Fall.