What is Machine Learning?

Machine learning, algorithms, artificial intelligence…these are all words you will come across as you train to be a data scientist. If you don’t have a background in computer science, coding, math, statistics, or anything from the sciences you may have only come across these words in your periphery or on the news. Now, I *like* math and coding, but I do not have a degree in them and, frankly, the way some of the concepts are described is always sterile and sometimes confusing.

In an effort to spread the love on data science, I’m going to try to tackle some of the most common concepts and keywords and re-explain them in my own simple words, without all the boring jargon. Join me.

So…what is machine learning?

To keep it short and sweet: machine learning is a bit of code with some math included (as in, there is code that does the math for you) that someone has written on a computer. Some historical data is then processed through this code and it spits out a prediction or decision, depending on what question you were looking to answer.

The data is always historical because we can’t predict the future, yet. The precise code and math will also vary depending on the programming language, or code, used and what type of question is being asked.

There are three main types of machine learning:

Note that I am intentionally not calling them by their “proper” names.

• One where you provide the “machine” (code written on a computer) with data that already has the answers to your question. The goal is when you add new data that it will assign the correct answer. Commonly used to predict house prices.
• Another is using data without the answers. The “machine” will group items in some logical way, that logic depends on the code. Commonly used to group different types of consumers based on their behaviours and other data.
• The third is one I have never used so I may trip up here…the “machine” can take data with answers and continuously learns from new data to make the best decision. It is what is used for self-driving cars — someone has given the machine data about roads and driving while the car continuously collects data from its surroundings to make an assessment.

That’s it, machine learning in the simplest terms!

If you enjoyed this and found it helpful (or totally unhelpful) please let me know by leaving a reaction or comment, or find me via jasminludolf.com.