Machine Learning Methods: Supervised Learning

In this article, we are going to review the most common and valuable machine learning algorithms which are frequently applied in the industry, academy and research. Note that some algorithms may have many subcategories or derivations since machine learning and artificial intelligence have been actively and extensively studied and utilised nowadays. This article will present a categorical overview of these fundamental algorithms and briefly explain each of them. In the succeeding articles, each algorithm will be deeply explained and their implementations will be exhibited. 

Machine learning (ML) algorithms are generally considered in two main categories as follows;

  1. Supervised Learning Methods
  2. Unsupervised  Learning Methods

Supervised Learning Methods 

Supervised learning techniques create logical connections or maps between input and output data. Therefore, these types of methods usually require a significant amount of labelled data for training. After that, these logical connections or maps created using training data can be utilised to evaluate or predict the output of a new data set. The performance of these types of algorithms is affected by the size of the training data-set such that having a larger data-set will probably lead to a better prediction accuracy on the unknown or new data. 

Widely Used Supervised Learning Methods

  1. Support Vector Machines (SVM)
  2. Linear Regression
  3. Logistic Regression
  4. Naive Bayes
  5. Linear Discriminant Analysis
  6. Decision Trees
  7. k-nearest Neighbor Algorithm
  8. Neural Networks 
  9. Deep Neural Networks (Deep Learning)
  10. Similarity Learning

Electromagnetic Modelling and Antenna Simulation via Opensource Software

Commercial electromagnetic simulation (EM) software packages such as CST Microwave Studio and ANSYS HFSS are widely used in commercial appli...