This is a multi-criteria decision-making problem. There are nine variables (criteria) and 50 options.
They are financial variables and I expect there are correlations between some variables. So I want to select the most suitable variables.
What method do you suggest for my problem?
Hey @kareem_nada,
Welcome to the community. When you use the term “variables”, do you refer to the input features? And when you use the term “options”, do you refer to the number of classes? In other words, does your problem involve 9 input features, and 50 classes to predict from, i.e., a 50-class classification problem?
Also, if so, then, by “selecting the most suitable variables”, do you refer to selecting the most important features? If so, then I guess your problem is a feature selection problem, and you can use algorithms like PCA (Principal Component Analysis), SVD (Singular Value Decomposition), etc. Often, you will find that algorithms that are used for dimensionality reduction are used for feature selection as well. You can find a comprehensive list of such algorithms here. I hope this helps.
Cheers,
Elemento
1 Like
I recommend you use all of the features. Correlation between the features isn’t a problem.
1 Like
Link of algorithms didn’t work
Hey @kareem_nada,
It’s working perfectly fine on my machine. Can you please try opening it in Incognito once?
Cheers,
Elemento