Summary of advanced feature selection methods - Jul 2022

1. SPSA-FSR: Simultaneous Perturbation Stochastic Approximation for Feature Selection and Ranking - An Introduction and Accuracy Performance Comparison with Other Wrapper Methods by Zeren D. Yenice, Niranjan Adhikari, Yong Kai Wong, Vural Aksakalli, Alev Taskin Gumus, Babak Abbasi [16 Apr 2018]
   • paper with code url
   • paper fulltext in pdf
   • implementaion with examples in python
2. PPFS: Predictive Permutation Feature Selection by Atif Hassan, Jiaul H. Paik, Swanand Khare, Syed Asif Hassan [20 Oct 2021]
   • paper with code url
   • paper fulltext in pdf
   • implementaion with examples in python

     • it is a wrapper-based feature selection method based on the concept of Markov Blanket (MB).
     • it can work for both classification and regression tasks on datasets containing categorical and/or continuous features.
     • according to the authors, the PPFS outperforms state-of-the-art Markov blanket discovery algorithms and well-known wrapper methods.