A Brief Introduction to Machine Learning-1

The aim of these series of articles is to give a brief introduction to machine learning. In the first article, our aim is to cover the followings:

1. Variable types

2. How are we learning?

3. What is machine learning?

Machines can predict the future, as long as the future doesn’t look too different from the past [1] .

1. Variable Types

While dealing with machine learning problems, it is crucial to be familiar with the variable types. In this section, we will explain some of the common terms.

· Numerical variable is a variable which can be an integer or a float.

· Categorical variable is also known as qualitative variable. Categorical variables can be further categorized as either nominal or ordinal.

Nominal variables are variables which do not have an order. For instance, being ‘male’ of ‘female’, football teams, etc…

Ordinal variables are variables which can be ordered or ranked. For instance, education status is an ordinal variable.

· Dependent variable is simply a variable whose outcome can be affected by the features. It is also called as a target or an output variable.

· Independent variable is a variable which can be manipulated to find out the effects to the outcome. It is also called as a feature, input, column, experimental and predictor variable.

2. How are we learning from a dataset? Could Leonardo Di Caprio survive?

We are trying to find some patterns between the data, and we are looking for which of the features plays an effective role on the outcome.

Let’s look at the picture in below. It is the first ten observations in the famous ‘Titanic’ dataset.

In this dataset, the columns ‘Pclass’, ‘Sex’ and ‘Age’ are our independent variables and the column ‘Survived’ is our target variable. The rows are called observations.

By looking at the first ten observations, we can come up with some insights:

· Male passengers in third Pclass could not survive. So, being survived and sex and Pclass can be related with each other.

· Male passenger in the first class also could not survive. But, on the other hand, he was 54 years old. Maybe, the age can be decisive for the male passengers in the first class.

· Female passengers survived. So, giving priority to women may have increased their chances to survival.

Therefore, it would be nice to examine the relation between the outcome and the features Sex and Pclass.

Unfortunately, by looking at these results, we can say that the raft was the only chance for Leonardo Di Caprio to survive… ☹

3. What is Machine Learning?

The machines imitate how human beings learn by using the data and the experience. They use the past data to find out patterns and insights. Then they try to predict the outcome of an unseen data.

Arthur Samuel defined machine learning as follows:

Machine Learning is the field of study that gives computers the ability to learn without being explicitly programmed.

Final Notes

This article is the first article of a series of machine learning notes. So, it is a gentle introduction to machine learning. I plan to go deeper to the machine learning algorithms in the upcoming articles.

I hope you find it helpful! If you have any questions, do not hesitate to ask it!

Refereces

1. https://www.miuul.com
2. https://medium.com/@randylaosat/a-beginners-guide-to-machine-learning-dfadc19f6caf
3. https://medium.com/analytics-vidhya/an-introduction-to-machine-learning-574bafa6fc66
4. https://towardsdatascience.com/what-are-the-types-of-machine-learning-e2b9e5d1756f
5. https://towardsdatascience.com/reinforcement-learning-101-e24b50e1d292