Olorunshola Tiwatope
4 min readJan 19, 2024


Machine learning is a typical phrase used in today’s world. Whether it’s in a book, on the news, or from the word of mouth of others, at some point, we have heard it. Well, if you haven’t, don’t worry. I’ve got you covered, and Machine learning can be ambiguous even to those who claim to know it. I will do justice to the idea to assist you in understanding what it is without involving the overly technical aspects.


Photo by Markus Winkler on Unsplash

Machine learning, sometimes called ML, is simply the process of training a system, e.g., not limited to computers, phones, IOT (Internet of Things) devices, and even robots, to be as intelligent as human beings. Intelligent as a human being? What do you mean? Human beings are the most intelligent species on the planet, and our intelligence enables us to perform various tasks, both simple and complex. It could be as simple as identifying friends and loved ones from a distance and as tough as driving on a very traffic-jammed highway (trust me, driving in Lagos can be crazy!). ML enables a system to do these tasks far better than humans.


Machine Learning Model Icon

A machine learning model is an algorithm — a set of instructions that learns patterns from data, enabling it to make decisions without it being explicitly coded.

The ML model is developed and trained on the system. Data, e.g. text, audio, videos, images, etc., is used to teach a model, and the system can identify patterns and trends from the data. For example, suppose we had a set of numbers such as {2,4,6,4,8,4,8}. In this case, we observe that the number “4” is the most occurring, and even further, the set contains even numbers.

Also, we have a set of text such as {cat, dog, rabbit, hamster}; we can identify that all these are animals and can also be pets. In the same way, we can find relationships within a set of data, so a system can also identify an individual’s faces in pictures and voices, even in audio. It can label objects or faces.

A machine learning model is an algorithm — a set of instructions that learns patterns from data, enabling it to make decisions without it being explicitly coded


Image from Potentia Analytics

The beauty of Machine Learning is that a model doesn’t have to be strictly programmed to do what you want it to. There are three types of ways to teach a model.

  • Supervised Learning involves feeding your model with data (input) and providing it with an expected output (data). For example, you give it a collection of audio labelled of two friends, Femi and Kayode. The model learns to identify the characteristics that make Femi and Kayode’s voices different. So, if either voice plays next time, an accurate answer of who’s speaking should be given.
  • Unsupervised Learning involves training a model with data and leaving it on its own to identify the patterns without any form of guidance. Consider a scenario where a company has a large dataset containing customer information but needs predefined categories or labels, e.g., age group. The goal is to identify natural groupings or segments within the customer base without prior knowledge of customer preferences.
  • Reinforced Learning involves an agent like a robot. The agent learns by exploring its environment; a reward system teaches the agent what to do and what not to do. It’s more trial and error by Learning; the agent discovers the best possible way of working towards achieving its goal. An example is a self-driving car left alone to drive on the road. There’s a finish line that it is to cross. Suppose it hits obstacles on the road; it will start again and repeat until it learns to avoid hitting the obstacles to reach its destination.


You will undoubtedly have seen Machine Learning in your everyday life, even if you don’t know it! The examples of scenarios of ML are

  • Self-driving cars
  • Facial, voice or biometric recognition for laptops, phones and security devices
  • Posts recommended to you when you use X, Instagram, LinkedIn, etc.
  • Netflix suggestions of movies and shows
  • Machine learning helps predict profit, business growth rate, and events.
  • Google Photos helps you classify pictures by face.
  • Robot Automation


Machine Learning assists with simple and complex problems. It also lets us see the future and helps us make decisions and act proactively, especially in businesses and organizations.

ML is becoming an integrated part of our lives; we can do much more regarding productivity and efficiency.