This repository contains machine learning algorithms developed by me and/or my team.
In this study, a novel machine learning algorithm for multiclass classification is presented. The proposed method is designed based on the Minimum Distance Classifier (MDC) algorithm. The MDC is variance-insensitive because it classifies input vectors by calculating their distances/similarities with respect to class-centroids (average value of input vectors of a class). As it is known, real-world data contains certain proportions of noise. This situation negatively affects the performance of the MDC. To overcome this problem, we developed a variance-sensitive model, which we call Standardized Variable Distances (SVD), considering the standard deviation and z-score (standardized variable) factors.
Main paper:
Elen, A., & Avuçlu, E. (2021). Standardized Variable Distances: A distance-based machine learning method. Applied Soft Computing, 98(2021): 106855. doi: https://doi.org/10.1016/j.asoc.2020.106855
In this study, an adaptive kernel is proposed based on the Gaussian function, which is used in Support Vector Machine (SVM). While the sigma parameter is determined as an arbitrary value in the traditional Gaussian kernel, the proposed method calculates an adaptive value depending on the input vectors.
Main paper:
Elen, A., Baş, S. & Közkurt, C. (2022). An Adaptive Gaussian Kernel for Support Vector Machine. Arabian Journal for Science and Engineering, (2022): 106855. doi: https://doi.org/10.1007/s13369-022-06654-3