Clocktimizer: An introduction to machine learning – Part 1
Nearly everyone who keeps up to date with technological developments has heard of the term ‘machine learning’. It’s a nice buzzword that is being used more and more in recent years. But what does machine learning actually entail? What is the technology that lies behind it?
To coincide with the launch of Clocktimizer’s machine learning engine, we will be releasing three blogs written by Susan Brommer. Susan is Clocktimizer’s resident AI genius and the person responsible for building much of our own machine learning engine. In the blog series we will explain what machine learning is, how it came to be, and what we can expect from it in the future. We will even go into some detail of the inner mechanisms of machine learning. In this first blog we are going to tackle what it actually is and why it is important to understand machine learning.
What is machine learning?
Machine learning is a form of artificial intelligence (another buzzword!). Artificial intelligence is pretty self explanatory. Namely, it is intelligence displayed by machines, rather than humans or animals. Machine learning in particular focuses on how computer algorithms can autonomously learn. These algorithms ‘learn’ in much the same way small children or smart animals learn; by interacting with the outside world.
Nowadays, machine learning algorithms are taught through a process of analysis of huge amounts data. It is important to realise that there is a lot of data out in the world. People send about 500 million tweets per day. Amazon sells 600 items per second. More than 200 billion emails are sent each day. All this data could be analysed using old school statistical methods, as those more grey haired of us will remember from our first year at university or college. However, it would be impossible for us to perform this statistical magic ourselves, simply because there is just too much data. Therefore we let the computer analyse the data for us.
By looking at all this information, a computer can come to conclusions about how the world works. It then can use that information to make predictions about the world. We humans in turn use these predictions to guide ourselves in the decisions we make. What is the next show you are going to watch on Netflix? What product will you buy on Amazon? How should you translate some word into another language?
Why should I care about machine learning?
Machine learning has entered our lives at a tremendous rate over the past few years. Probably the best known examples of this are private corporations like Google, Amazon, and Facebook. These companies use machine learning to personalise their services for their customers, but also to predict and steer the behaviour of their customers. Increasingly, with as many negative ramifications as positive.
Universities all around the world are investing in machine learning. Not only from a computer science perspective, but also within other research fields. Social sciences, for example, has been using machine learning to find out whether social media can predict election results. Clearly, following the predictions of Hilary’s guaranteed success in 2016, they could use all the help they can get. Even governments invest in machine learning and apply it in various ways. Among other things, they use machine learning to detect tax fraud, analyse crime, and automatically analyse security camera footage.
This might already give you a good idea of how machine learning affects your life, but it gets even more personal. Banks use machine learning to determine your credit limit. Social media companies use it to filter content based on your behaviour. Did you ever notice that online you get different advertisements to your friends? That is because machine learning algorithms personalise marketing for you. In short, machine learning is too large to ignore. Both from a personal perspective and from a business perspective.
How should I use machine learning?
Machine learning is a fast moving trend. There’s a big chance that you want to hop on the train before it passes you by. But how exactly should you use machine learning? Since machines learn from data, you can answer this question by investigating what data exists within your company (or what can be created). First ask yourself: What data do you have? What data is easy to collect? What data is less easy to collect, but may contain interesting information? With the rise of machine learning, information has become a valuable asset. It is never too late to start collecting information about anything related to your company.
Next, find out what information you would like to acquire by analysing the data. Typically, you want information that helps you guide your decisions, or helps you predict the future. Start with finding out what valuable information is hidden in your data. Then ask yourself what you want to know more about. Make sure you have a goal for the machine learning algorithm. Go back to the data collection step if the data does not fit your goal.
Lastly, let the machine do its job. Either hire a data scientist that can start the process, or use a tool like Clocktimizer, which can do the machine learning for you. After this step, you might refine the process a bit. Tweak what data you are collecting. Or come up with an entirely new goal for data analysis.
What to expect in Part 2 & 3
If you want to know more about what machine learning has already accomplished, what kind of machine learning algorithms there are, and how they work, then make sure you read the next two blogs of this series! We’ll be sharing them in the coming weeks as part of our ongoing #NoCodeNoCry campaign.
