Introduction to Machine Learning Algorithms

3 min readAug 11, 2023

~Hola folks!!! It’s been a long time… Welcome back with another post 😊…

First, let’s see what Machine Learning is…

Machine Learning (ML) is a subfield of Artificial Intelligence which enables computers to learn without being specifically programmed. ML allows machines to learn on their own. Moreover, ML consists of Algorithms (Heuristics) and Data-driven methods.

Although Machine Learning has vast amount of capabilities, it also has limitations. Therefore, ML won’t be accurate all the time.

There are 4 Types of Machine Learning Algorithms.

Supervised Learning
Unsupervised Learning
Semi-Supervised Learning
Reinforcement Learning

Supervised Learning:

Supervised Learning supervise when training models with use of labeled data. It gives input data to the model along with the output. There are 2 types of supervised learning algorithms. They are, regression and classification.

Numerical data will be analyzed in regression problems whereas binary data and categorical data will be analyzed in classification problems.

Linear Regression, Decision Tree and Support Vector Regression are some of the regression algorithms. Logistic Regression, Decision Tree and Support Vector Classifier can be used as classification algorithms.

Unsupervised Learning:

Unsupervised Learning find the structure and patterns from the input data which are not labeled. This is used to group data into clusters which are then utilized in decision making.

There are 2 types of unsupervised learning algorithms. They are, clustering and dimensionality reduction.

K Means clustering and Density Based Spatial Clustering of Applications with Noise (DBSCAN) are some of the clustering algorithms. Principal component analysis (PCA), Multi Dimensional Scaling (MDS) and Linear Discriminant Analysis (LDA) can be used as dimensionality reduction algorithms.

Semi-supervised Learning:

Semi-supervised Learning involves a small portion of labeled data and a large number of unlabeled data since labeled data is expensive and difficult to get while unlabeled data is cheap and easier to get.

Generative Adversarial Networks , Auto-encoders and Variational Auto Encoders are are some of the semi-supervised algorithms which are basically used for data manipulation.

Reinforcement Learning:

Reinforcement Learning is a self-teaching system that essentially learns by trial and error. Reinforcement learning includes algorithms that learn from results and decide which action to execute next. After each step, the algorithm receives feedback that allows it to determine if the choice it made was correct, neutral, or incorrect.

There are 2 types of reinforcement learning algorithms. They are, positive reinforcement and negative reinforcement. Positive Reinforcement takes place when an event occurs as a result of a behavior and enhances the strength and frequency of the behavior, whereas Negative Reinforcement occurs when a negative scenario is stopped or avoided.