In our previous discussion we said that machine learning algorithms are not difficult. However, we need to know mathematics, and to be precise some concepts of discrete mathematics.
And besides, we need to know how we can write algorithms in Python.
Why are we saying so?
Let’s try to understand.
Because one of the main branches of discrete mathematics is also algorithms.
That is why we use concepts and notations from discrete mathematics when we write algorithms in Python.
What do we actually do?
We just translate a mathematical problem to the python algorithms.
Why?
Because, in mathematics, we solve the same problem using algorithms.
What are algorithms?
We’ve learned before. It’s a recipe. A step by step guide.
That’s why, when we combine mathematics and python, it becomes machine learning algorithms. However, when we add our domain knowledge with it, it becomes data science.
It’s as simple as that.
Let’s take a look at the following image.
The image says a thousand words. Right?
Therefore we can combine mathematics and python and get the necessary machine learning algorithm we need for data science
Let’s try to make it easy so a beginner in data science can understand what we want to say.
We know that mathematics is much older than computer science. In short, mathematics is much older than python.
However, from day one of mathematics, algorithms were there.
Why?
Because mathematics also needs clear and concise instructions.
In other words, algorithms should represent clearly defined meaning.
Yes, we are talking about semantics. We have learned what syntax and semantics is important before.
A distinct meaningful output should come out from the inputs.
Mathematics and algorithms
The question is why did we need algorithms four thousand five hundred years ago in Babylon?
Why did we need it three thousand five hundred years ago in Egypt, or later in Greece?
The answer is: to decide something. When we travel by one car and come to a road-divider that indicates two ways, we cannot go to two ways simultaneously.
Our decision should be discrete. 1 or 0.
True or false.
Which is discrete mathematics, a branch of mathematics.
In contrast, if we have two or more cars, the decision might be something else.
Although ancient Babylonian, Egyptian or Greek mathematicians started using the concepts of algorithms since antiquity.
The term ‘algorithm’ is derived from the name of the ninth century Persian mathematician Muḥammad ibn Mūsā al-Khwārizmī.
Although much before that, Greek mathematicians used the sieve of Eratosthenes to find prime numbers; they also used the Euclidean algorithm to find greatest common divisors (GCD).
We’ll learn them step by step.
Therefore, let’s extend this topic to the next discussion.
So stay tuned.
Leave a Reply