Say, there is a recently proposed machine learning algorithm that you applied to another dataset and you found high performance results.
In general, you found a method that was proposed to solve some task and you apply it to solve another task.
Is this enough for a contribution? Isn't this plagiarism?
It is not plagiarism as long as you properly cite the paper. There are tons of papers titled as follows [task] of [dataset] using [method]. For example, A, B and C (2077) Estimating parameters of a stochastic cell invasion model with fluorescent cell cycle labelling using approximate Bayesian computation. Journal of the Royal Society Interface. xx, xxxxxxxx.
If this is classified as an act of plagiarism, then most research papers are just a product of plagiarism. We more or less stand (rely) on the shoulders of "giants".
It's not plagiarism.
However, if you submit it to a good journal or good conference, it will likely be rejected as not being sufficiently novel for their journal/conference. A sufficiently low-ranked (but still legitimate) journal/conference will accept it.
(Notice: Novelty is NOT a binary variable.)
