Tuesday, October 4, 2016

Machine Learning 1 - What is machine learning and real world example

From this week, I will try to explain some basic machine learning concepts and algorithms in a more intuitive way without equations. Note, all these will based on my knowledge of machine learning, therefore, it can be inaccurate or even wrong, feel free to contact me if you find anything wrong, or have a better way to explain. 
Let's start today with the following few questions. 

What is Machine Learning (ML)?

As the name suggests, ML is the methods that enable machines or computers to learn from data we give them and then take decisions/actions. The famous definition from Tom Mitchell's book - 'Machine learning' is more formal if you want it.

To add a simple example, if we give thousands of images with cats or dos on it, and let the computer to learn from the characteristics of cats or dogs. After learning (of course, it is the machine learning), we give some new images to the computer, and ask it to decide whether it is dog or cat. This is an example of what is machine learning. We will give more examples about different types of machine learning next week.

Real world ML examples

Think about some daily stuff we take for granted, you will find ML is everywhere: 

1. Why ATM can recognize your handwriting on your check?

2. Why Amazon can recommend me with good books?

3. Why the camera can detect my face?

4. Why some emails be put into spam folder?

5. Why my bank call me about fraud?

6. Why self-driving car can drive by themselves?

7. Why google search engine can return what I want to search?

8. Why voice assistants (Siri) on my phone understand me?

9. Why advertising can target me correctly?

10. Why home robot is so smart?

All the figures are from internet, I thank the authors for the figures!


  1. Replies
    1. 哈,好久不见了,很久没听有人叫我这么亲切的名字了。你最近可好?

  2. Very quickly this website will be famous among all blogging people, due to it’s fastidious articles or reviews
    Separating machine manufacturers

  3. Fantastic post! Would you mind if I used the first picture in an upcoming ML presentation?

    Thank you!

    1. Hi Andrew, feel free to use the figures, all the figures I got from internet and I have the last sentence of the credit "All the figures are from internet, I thank the authors for the figures!"

    2. Thank you, Qingkai! Keep up the great work!