An Overview of Machine Learning algorithms

Here we would like to give a short review of some well-liked Machine Learning (ML) algorithms. Nowadays, there exist so many algorithms but how you can select the right one for your business growth? In any field, you may work, you require to categorize the algorithms in the following ways:

• Grouping algorithm by their learning style;
• Grouping algorithm by their similarity in form or in function;

Now before going deep into the algorithms let's discuss what is Machine learning algorithms? As we may know ML algorithms are the brains in any model which helps machines to learn and make the machines smarter. The way the ML algorithms perform is that it is equipped with the batch of data and over time these algorithms improve their accuracy. By accuracy, it means how much the predicted value is precise.

The taxonomy of ML algorithms is necessary since it helps us to understand the role of input data and the modeling process. During those processes, it is crucial to select the proper algorithm to obtain the optimum results.

Here are three different learning styles in ML algorithms:

• Supervised Learning: Input data is the training data and has assigned a label. Examples of this style are classification and regression problems. And algorithms are logistic regression and Back Propagation in the neural networks.
• Unsupervised Learning: Input data is Not labeled and it may extract general rules such as mathematical processes to reduce redundance in data or organize data by similarity. Examples of this style are dimensionality reduction techniques, vector quantizations, clustering methods. And algorithms are K-means, Principal component analysis (PCA).
• Semi-Supervised Learning: As it is clear from the name it is a mixture of labels and not labeled input data.

Now regarding the algorithms that are grouped by similarity w.r.to their functionality like how they operate. Although there are still some algorithms that can fit into multiple categorize such as Learning Vector Quantization (LVQ) methods that are both neural network method and instance-based one.

• Instance-based?Algorithms: such as k-Nearest Neighbor (kNN), Learning Vector Quantization (LVQ), Self-Organizing Map (SOM), Support Vector Machines (SVM).
• Regularization Algorithms: such as Ride regression and Elastic net.
• Regression Algorithms: such as Linear regression, logistic regression. stepwise regression.
• Decision Tree?Algorithms
• Bayesian Algorithms
• Clustering Algorithms
• Artificial Neural Network Algorithms: such as perceptron learning, Multilayer perceptron (MLP), Back-propagation, Stochastic gradient descent, Hopfield Network.
• Deep Learning Algorithm: such as Convolutional Neural Network (CNN), Recurrent Neural Networks (RNNs), Long Short-Term Memory Networks (LSTMs).
• Dimensionality Reduction?Algorithms: such as Principal Component Analysis (PCA), Principal Component Regression?(PCR), Partial Least Squares Regression (PLSR).

There exist other lists of ML algorithms and if you are interested please follow the below link for your kind review:

ML algorithms