I am looking to use different machine learning approaches to build some regression models. if it will possible to use machine learning methods for regression purposes?
I have small genetic and phenotypic data (around 1000 samples)?
Hello,
generally there are many regression methods that can be used in your case, the simplest one can be the PLSR or Partial Least Squares Regression method, also some ML methods such as Random Forrest, ANN...Etc,
With 1000 samples and depending on the correclations of your data, I can advice you to use the PLSR.
Good luck.
*Issam*
Dear Mohamed Belhocine ,
Of course it is possible, you can find your methods in this link:
https://www.analyticsvidhya.com/blog/2021/01/a-quick-overview-of-regression-algorithms-in-machine-learning/
Best,
Dear Mohamed Belhocine
Precisely yes, these articles might be useful further, have a look:
1. https://res.mdpi.com/d_attachment/sustainability/sustainability-13-08600/article_deploy/sustainability-13-08600.pdf
2. https://link.springer.com/article/10.1007/s10994-019-05848-5
3. https://www.nature.com/articles/s41398-020-00957-5
4. https://www.sciencedirect.com/science/article/pii/S0176161720302443
5. Article Eye-color and Type-2 diabetes phenotype prediction from geno...
6. Article Deep learning versus parametric and ensemble methods for gen...
Kind Regards
Qamar Ul Islam
yes of course you can. For regression methods, there are many methods that depend on the type of data, if your data are characterized by a linear distribution, you can use the multiple linear regression, the principal component regression, or the partial least squares regression. If the data are not linear, you can use nonlinear regression methods such as ANN (Artificial neural network), or the SVR (support vector regression)
- Badges
- Science topic