Automatic english grammar correction is a complex problem that requires advances in multiple discplines like language modeling and machine learning. However, a native english speaker (given enough amount of time and concentration) can correct and possibly enhance any piece of text in exchange of a monetary reward, and when many such individuals collaborate simulatenously they can even exceed the performance (speed) and match the quality of an expert english proofreader. The Soylent paper[1] introduces the idea to proof read text with the help of the crowd, in this project we aim at minimizing the amount of requests sent to the crowd by trying to find the correct senstences.

[1] Bernstein, Michael S., et al. “Soylent: a word processor with a crowd inside.” Proceedings of the 23nd annual ACM symposium on User interface software and technology. ACM, 2010.