You’ve heard of Artificial Intelligence (AI) and now you likely keep hearing about Machine Learning (ML). Are they the same?
The short answer is no. You can think about AI as an overall field of study and ML as a subset of it. AI is the science of trying to get a computer to mimic the human mind with the ability to do a task with reason. ML is an application of AI that gives a machine access to data and the ability to learn from that data (no reasoning needed).
So how does ML work? Let take a very simple example: Fruit.
(Note: This is an extremely simple example and does not cover all the details of ML strategies and designs.)
How can you tell the difference between an apple and a banana? Color is a good starting point. How about skin hardness? Or shape. Over time, we collect data about all types of fruits to help us more quickly identify fruit. In other words, the more data we collect about the fruit, the better the “machine” can learn. Similarly, the success of ML is dependent on big data. The more data we can provide, the more accurate the ML can be.
Next, we get a bunch of fruit together. This will be the test data. We give the computer the test data and the information we collect about colors, shapes and skin hardness. We let it learn. The computer starts looking at the fruit and the data collected, and it learns to determine what is a Granny Smith apple and what is an unripe banana. From this process, it outputs a model. This model can now be added to a program that can identify new pieces of fruits.
Ok… that’s a simple example, but how can it help me in the real world of business?
Well, I would say you probably are already being helped by ML and don’t even know it. Think about your email system. Over the years, it’s been getting better and better at filtering out spam emails. Hasn’t it? What has made it better? For starters, ML has been applied to email filters and it has been learning how to detect spam emails faster and more accurately.
Here are a few other examples:
- When you talk to Alexa, it uses ML to figure out the question you are asking using a form of ML for natural language processing.
- Have you used Google Translate? It uses ML to help with language processing.
- Ever wonder how Facebook knows who you and your friends are in a new photo you posted? It’s a form of ML that uses computer vision to figure out who is who.
- Have you used grammarly.com? They use ML to help you compose email without the grammar mistakes.
These are all examples of how ML can help us today. Computers have become faster and faster over the years and now have the ability to process large amounts of data quicker than we humans. This allows machines to help us make better decisions.
Your computer will not become self-aware and annihilate you (at least not in the near future), but it will help you process more information in a timely manner that will help you make better decisions and increase your business productivity.
You may also like from Scott:
- Internet of Things: How Are You Adapting?
- Progressive Web Apps (PWA): Another Tool for Mobile Development
- The Developer’s Guide to Mobile Application Development