0
$\begingroup$

I may want to ask if there are studies that exist which utilize a "three-point machine learning model. What I mean by "three-point machine learning model is that it may use several algorithms in order to construct a model (in this case, it would be three machine learning algorithms).

An example would be coming from our study which utilizes Random Forest (RF), Gaussian Mixture Model (GMM), and Radial Basis Function (RBF) Neural Network in order to construct a machine learning model within the context of multi-label classification.

I may just want to ask also if a "three-point approach" is a feasible way to construct machine learning models like what we have in mind?

Improve this question
$\endgroup$
6
  • $\begingroup$ Are you asking about stacking? $\endgroup$
    – user2974951
    Commented Mar 23, 2023 at 16:28
  • $\begingroup$ not exactly stacking, but utilizing several algorithms to build a model (for example, the process of the model involves feeding data to RF algorithm for dimensionality reduction, then the outcome value of it would be then fed to GMM for feature learning, the outcome value of it is then finally fed to RBFNN for classification. $\endgroup$
    – Ralph Henry
    Commented Mar 23, 2023 at 16:33
  • $\begingroup$ Yes, it is very common to apply different algorithms on each stage and chain them together. In fact, hardly any production system does not do this. $\endgroup$
    – lpounng
    Commented Mar 24, 2023 at 4:46
  • $\begingroup$ Depending on how the models are chained, it may be called a pipeline (e.g. your example), or stacking (bagging, boosting) etc.. $\endgroup$
    – lpounng
    Commented Mar 24, 2023 at 5:01
  • $\begingroup$ You may find a lot of examples in sklearn's doc. $\endgroup$
    – lpounng
    Commented Mar 24, 2023 at 5:02

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.