Where is Machine Learning not used?🤖 Machine Learning is an integral part of our everyday lives. Because of the digital transformation and computerization in the last years, we have adapted to this new digital era. Therefore, we use machine learning every day; why? Because we want to learn from data rather than compute hardcore solutions.
How this happened too fast? Well, computers became cheaper. Forget those times when computer were rather big calculators. Yeah, big calculators which processed numbers. But numbers represented data, hence, computers were, even at that time, able to process all kinds of data (Figure 1).
"101100" Represents either the decimal 44 and the comma.Figure 1. Binary digits
Should you be aware of Machine Learning?
Yes, you do. We are constantly taking benefit from Machine Learning Algorithms. For example, when getting recommendations for movies, items to buy and modern smartphones, facial recognition and image recognition in your cloud of preference.
As a Data Scientist, I believe it is always fundamental to master the basics (Concepts and programming). Thus, let's try to understand the concepts of Machine Learning.
Basic types of machine learning:
There are two main types of machine learning problems, supervised learning & unsupervised learning. There is, to a less extent, the reinforcement method.
Supervised Learning
In supervised learning, the basic type of machine learning, the algorithm is trained with a properly labelled data set.
Yeah, I know, you have to clean data, always.
Which kind of data? Images with people!! How many times have you watched your phone and labelled the person in the picture? Well, your small computer (Your phone 📱) applies an algorithm that enables it to recognize this labelled person in the next captures. -Yeah, I know, it's scary.
Be aware about the data you share, even names!
Unsupervised Learning
In this kind of algorithm, you are not required - yes, human interaction is no needed. How can the algorithm then find the exact relationship in, for example, images?, Creating structures. These structures are data points that make the algorithm versatile and can adapt dynamically by changing the structures.
Do not worry, we will dive into these machine learning algorithms for better understanding.
Reinforcement Learning
I would describe this type as a trial and error method. Basically, how humans behave and learns. It learns and adapts to new situations. Reinforcement learning predicts a sequence of actions. Reinforcement learning is connected to a reward.
A classic example of reinforcement learning is, how to play a video game, either win or lose, you learn!!
How is machine learning related to artificial intelligence?
Machine learning is a subset of AI. Basically, Machine Learning is a branch of artificial intelligence that enables machines to learn by example.
Computers are learning more because more data and more computer power are accesible.
Before we dive into coding, let's talk about some Machine Learning examples.
At the airports, for facial recognition.
Computer vision.
Biometrics.
Face detection.
Audio and Text Recognizition.
That's it, we covered the concepts of Machine Learning. In the next posts, I will dive into code and the classic Machine Learning tools you need to become a Data Scientist. 👉 Follow me?
Do you know more examples? Please give me a clap, follow me and comment on the post. Invite me a coffee! 👏
Check my channel for machine learning tutorials.
