I'm conducting research on cyber bullying classification. I recently found a github repository containing code on sentiment analysis which looked at different machine learning algorithms, doc2vec + ml classifiers and a word2vec + cnn model. My dataset is different and I do use different machine learning classifiers than the github code does. However, the way the code in the repository trained the cnn model and the word2vec that was used is exactly the same.

Would using the code in the repository for my problem be considered plagiarism?

  • 3
    The question is a bit confusing: how does the title relate to the text? Are you planning to use the existing code on GitHub, or is your implementation doing the same thing as the existing code? – GoodDeeds Aug 15 at 1:09
  • basically, the github code has three parts which i will be using but slightly tweaking the code – user128373 Aug 15 at 1:13
  • 8
    Does this answer your question? How do you cite a Github repository? – GoodDeeds Aug 15 at 1:19
  • 1
    As long as you cite the source it's not plagiarism. The question of whether it's an original enough contribution is a different matter. – Erwan Aug 15 at 13:49
  • @Erwan I think you could post that as an answer. – Federico Poloni Aug 15 at 14:52

It would be plagiarism only if you don't cite the source. There's no plagiarism issue if you properly explain that you use this source code as a basis for your implementation and make it clear that it's not yours (preferably by citing a paper from the original authors if any, if not just give the github link).

Whether this will be considered an original contribution by the reviewers is a different matter. It might be ok if your motivation and data are significantly different from what the original authors did.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.