Along with these : https://machinelearningmastery.com/feature-selection-with-real-and-categorical-data/

Current a feature based regression technique is also being used

Dear Iman Javadzarin

Look the link, maybe useful.

https://towardsdatascience.com/the-5-feature-selection-algorithms-every-data-scientist-need-to-know-3a6b566efd2?gi=3dd28f4b3576

Regards,

Shafagat

Hope this will help you https://analyticsindiamag.com/what-are-feature-selection-techniques-in-machine-learning/

Well, it depends on your specific problem. For instance, whether you are going to address a regression problem, I recommend you start with computing Pearson's correlation between all the features analyzed. Otherwise, if your problem corresponds to a classification problem, you can use a recursive feature elimination algorithm. This algorithm is within of scikit-learn library, and it is easy to implement with a few code lines. Good luck!!

Dear Iman Javadzarin

I recommend you Genetic Algorithm, as i used it personally it worked outstanding.

Article Multi-Person Tracking and Crowd Behavior Detection via Parti...

Dear Javadzarin,

The following paper could help you

Article Improving nature-inspired algorithms for feature selection

Try Multi-Layer Sequential Forward Selection (MLSFS) algorithm proposed in one of our articles. It caters to the problems of SFS Algorithm.