Natural Language Correction-Thesis Proposal

Tools that correct errors in human written texts are an important part of many existing systems used by countless people. Although the rise of deep neural networks in recent years made it possible to handle more complex errors including even grammatical and fluency ones, current models still suffer from several problems. In this thesis proposal, we describe issues we find are the most crucial in the area of the natural language correction and discuss what has been done so far to solve them. We further describe the work we performed until now, and finally we provide an overview of our future research plans.