A short explanation would be:
“Machine learning is a type of artifical intelligence (AI) that provides computers with the ability to learn without being explicitly programmed.”
Using algorithms that iteratively learns from data, machine learning allows computers to find hidden insights without being explicitly programmed where to look. They learn from previous computations to produce reliable, repeatable decisions and results.
Why the increased interest in machine learning?
The data volumes are exploding, more data has been created in the past two years than in the entire history of the human race. On the other hand, computational processing is getting cheaper and more powerful with affordable data storage.
All of these things means it’s possible to quickly and automatically produce models that can analyse larger and more complex data to deliver faster and accurate results. These results or high-value predictions can guide business to make a better decision and produce smart actions in real time without human intervention.
Where is machine learning used today?
A list of some of our day-to-day activities that are powered by machine learning algorithms:
- Fraud detection
- Web search results
- Real-time ads on the web pages and mobile devices.
- Email spam filtering
- Pattern and image recognition
What are the machine learning methods?
There are two most widely adopted machine learning methods which are supervised learning and unsupervised learning. Most machine learning is supervised learning, there are some other type as well that are Semi-supervised and reinforcement learning.
Supervised learning is where you have input variables (x) and an output variable (Y) and you use an algorithm to learn the mapping function from the input to the output.
Y = f(X)
The goal is to approximate the mapping function so that you have new input data (x) and you can predict the output variable (Y) for that data. Because it is supervised learning we know the correct answers and the algorithm iteratively makes another prediction and minimise the error. The learning/ iterative steps stop when the algorithm reaches an acceptable level of performance.
Unsupervised learning is where you only have input data (X) and no corresponding output variables. The goal of unsupervised learning is to model the underlying structure or distribution in the data in order to learn more about the data. There is no correct answer so the algorithms are left to their own to discover and present the interesting structure in the data.
An example would be if thousands of unlabeled handwritten numbers are given to the machine in can use one of its method of clustering. Where you want to discover the inherent grouping in the data, such as in our example grouping all the 1 digit together, the 2 digit in its own group and so on. Another example could be grouping customer by purchasing behaviour that your business can focus and target.
I hope this give you a brief introduction to what is machine learning.