Machine Learning

 

Hello and welcome back to my blog !

In this blog I am sharing some important basics of machine learning. Machine learning can be confusing, so it’s important that we begin by clearing defining term : 

   

Machine learning is a type of Artificial Intelligence (AI) that allows a software program to become more accurate at predicting outcomes.  Or Machine learning is an application of AI that enables system to learn and improve from experience without being explicitly programmed.

For example, an algorithm would be trained with pictures of cars and other objects, and machine would learn ways to identify pictures of cars on its own. In that case we can say that our machine is learning or predicting outcome without explicit programming.

Machine learning relies on input, such as facts and figures, graphs, domains and connection between them. Machine learning starts with the observation of data, such as examples, direct experience or instruction. 

The term “Machine learning” was coined by Arthur Samuel, a computer scientist at IBM and a pioneer in AI. He designed a computer program for playing checkers (that I have already mentioned in my previous blog “Artificial Intelligence: An Exordium”). The more the program played, the more it learned from experience, using algorithm to make predictions.

Machine learning is not a science fiction. It is already widely used by businesses across all sectors to increase productivity and efficiency. Data security, Finance, Healthcare, Fraud detection and Retail are some major fields which are actively using Machine learning.

Machine learning depends on number of algorithms for turning a dataset into model. Which algorithm works best depends on the kind of problem we are solving, available resources and the nature of data. Followings are the most common algorithms used in machine learning :

• Regression algorithm.

• Classification algorithm.

• Clustering algorithm.

• Dimensionality reduction algorithm.

• Optimization methods.

Machine learning is not a new stream but we need to know that it is not perfect yet. Machine learning is still in developing phase and is still critical to the success of AI. But one thing is certain that together AI and ML (aka Hybrid AI) are changing the future of our machines.