What Is Machine Learning
A Very Basic Explanation
You may have heard the term lately and there is a lot of momentum gathering pace for machine learning. Due to a number of reasons including the vast amounts of data being generated by basically any connected device and companies and users are looking for different ways to use that data.
When you think of machine learning though, don’t think about artificial intelligent robots hunting you down because they have found a reason to kill all human beings. Machine learning is happening right now and helping with things like facial recognition software, data mining and making sure you are getting what you need from your google searches, just to name a few.
But what is machine learning?
If you’re familiar with traditional programming, we write code which will read in data. The control logic of the program will determine the correct parts of the application to execute depending on the specific data that is accepted. For a traditional program to be smarter and recognize even more of your data, we then need to program in more logic into our program.
This is where machine learning is different.
In machine learning we gather data and set it up into specific formats which machine learning can use. This means that we pass this information to an algorithm which will analyse the data and create a model to then implement a solution based on the data that we have passed in.
If we looked at a small set of data, we would be able to find patterns in the data which might point us to answer a specific question. For example, if you wanted to know the probability of a car breaking down in the next month. If you had data on different aspects of the car, for example, the age of the car, when was the last time it was serviced, when did it last have its oil changed, how many kilometers the car has driven.
From all these data points you could then give a probability of whether or not a car would break down in the next month or not. This is where machine learning comes in. In today’s world, the amount of data points for a single item, like a car, is becoming too unworldly and large for a human to analyse, combine this with the fact that the number of cars continue to increase daily. This is where a machine can work through this data and find the patterns much faster than a human could.
You need to remember though that the the goal of machine learning will never be to make a 100% correct prediction of probability. The goal of machine learning is to make predictions that are good enough to be useful.
Found this post useful? Kindly tap the ❤ button below! :)
About The Author
