Exploring Plagiarism Detection using NLP techniques

I am using NLP to detect plagiarism in written text documents. I will be using text preprocessing, shallow and deep NLP techniques along with different similarity metrics for similarity comparisons. After that I will of course use a ML classifier to classify the text as plagiarised or non plagiarised. I am using 'A Corpus of Plagiarised Short Answers' as the dataset which contains both suspicious and source text passages. I have currently started coding the preprocessing.

I need help mainly with the coding and the NLP aspect of my project. I would also like to discuss the project for help and suggestions regarding where the project can go or how to improve it. I can provide a more detailed description about the framework for the plagiarism detection I am planning to use.