I have received comments from reviewers on a paper regarding applying ML techniques (RF, KNN, Nnet etc) on a data with 384 samples. One of the reviewers is extremely concerned that about using machine learning techniques for 384 samples as it seemed very small to him/her. If there is any paper where ML is used for samples of 350 or less, it would be greatly helpful for me in this regard. I could cite the paper as an evidence that ML can be used for relatively smaller samples.
TIA.
I don't know of any papers, but 350 samples doesn't seem very many. Even in normal statistics, if you take 350 random samples from a normal distribution with one variable, the standard deviation of the mean of the samples is about 1/20 of the standard deviation of the true population, and the shape of the distribution may not even look very normally distributed. I would expect machine learning to take a lot more samples.
Dear researcher, what was your database? There are public data platforms such as Moodle, Massive Open Online Course (MOOC) that have a large amount of data. Observe the works of researchers who used such platforms and observe the minimum data used by the respective surveys. However, I agree that the sample is small in terms of data mining.
Article Application of SVM and KNN to Duval Pentagon 1 for transform...
Check https://bmcmedinformdecismak.biomedcentral.com/articles/10.1186/1472-6947-12-8
Also check https://www.sciencedirect.com/science/article/pii/S2468502X17300086
https://link.springer.com/article/10.1007/s42979-021-00592-x
Check these two papers:
Article Using environmental variables and Fourier Transform Infrared...
Article Predicting weathering indices in soils using FTIR spectra an...
Hi Maryam,
It may dependent on your variable nature, for example in my study case 114 samples seemed almost enough when RF used.
https://www.sciencedirect.com/science/article/abs/pii/S2352009421000560?via%3Dihub
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size