DEV Community


Machine Learning Glossary for Beginners

Rapinno Tech
We are software developers.
・2 min read

Machine Learning (ML), is the field of study that relates to computer algorithms that are capable of improving automatically though experience. Over the last few years, ML has gained prominence due to the cutting-edge applications in diverse sectors, such as aviation, healthcare, and finance. This has motivated many to delve into the field.

This article covers the basic terms for beginners so that you will have a strong base of knowledge to support you in learning advanced ML concepts.

The 4 types of Machine learning systems are:

Supervised learning

This includes providing numerous labelled data for the algorithm to learn and generate underlying rules.

Unsupervised learning

A large amount of data is provided to the algorithm to identify patterns and categorize them based on observed similarities.

Semi-supervised learning

As the name implies, it uses limited labelled data to produce underlying criteria. Based on this, numerous unlabelled data is provided for learning.

Reinforcement learning

It is a trial and error method. The algorithm is provided with a large number of unlabelled data, processing which will grant it positive or negative feedback based on the result.

Having covered the types, we can proceed to learn the basic terminology you will routinely come across in the realm of ML.


Natural language processing essentially teaches computers human text and speech and operates using them.


Mark 1 Perceptron was the first-ever custom-built machine for image recognition developed by IBM, which failed to recognize many patterns. The software Perceptron was salvaged to be used in the future.

ANN- Artificial Neural Network

A set algorithm is modelled with feedback loops similar to our neural system. Many layers of these neurons are used to create a Neural network or Artificial Neural network. It is used to solve complex problems. Multi layering helps to exponentially improve the processing power of this system. Each layer can be used for distinct data attributes.

The most significant ML algorithms are:


The Nearest neighbour algorithm that is used in pattern recognition


It helps hidden neural layers to adjust to a newer set of characteristics presented by the data.


This makes use of many weak learning algorithms to produce a strong learner.

The field of machine learning is vast and exciting. There is still a lot to learn through reading and experimentation. We hope this article has introduced you to the crucial fundamentals and has inspired you to take your learning to the next level.

Discussion (0)