Hello everyone, this article will be a part of a series of articles on Machine Learning. In this
series, we will take you from the basics of Machine Learning and teach you a few of the famous
algorithms used in Machine Learning. This article will be a soft introduction for those of you
who are not at all familiar with this particular field. So let’s dive in.
What is Machine Learning?
There isn’t a well-accepted definition for Machine Learning but one definition for the term
given by Arthur Samuel in 1959 is “Field of study that gives computers to learn without
explicitly being programmed”. An example of such a task would be automatic spam detection.
Another well-known definition of Machine Learning is “A computer program is said to learn
from experience ‘E’ with respect to some task ‘T’ and some performance measure ‘P’, if its
performance on ‘T’, as measured by ‘P’, improves with experience ‘E’.”
There are basically two major types of Machine Learning:
- Supervised Learning: With Supervised Learning, we have a labelled set of data with both
positive and negative examples of what we want the Machine to learn to predict and we
train the Machine on that dataset. It’s basically learning from experience.
Supervised Algorithms are further subdivided into two categories:
> Regression Problems
> Classification Problems
- Unsupervised learning: Unsupervised learning allows us to approach problems with
little or no idea what our results should look like. We can derive structure from data
where we don’t necessarily know the effect of the variables.
Unsupervised Algorithms are further subdivided into two categories:
> Clustering Algorithms
> Non-Clustering Algorithms
- Semi-supervised learning: Semi-supervised learning is a class of supervised learning
tasks and techniques that also make use of unlabeled data for training typically a small
amount of labelled data with a large amount of unlabeled data
The figure below should help you understand some of the definitions and terms mentioned
above in the article.