Machine learning is a subfield inside Artificial Intelligence and the most dominating one in current scenerio.
IBM describes Machine learning as -
"Machine Learning is a branch of AI and Computer Science which focuses on the use of data and algorithms to imitate the way that humans learn , gradually improving the accuracy."
Another cool definition that I found on the internet is -
"It is a branch of AI that enables computers to "self-learn" from training data and improve overtime without being explicitly programmed."
Machine Learning is further categorised as -
1.Supervised Learning
2.Unsupervised Learning
3.Reinforcement Learning
Out of all the economy that ML has generated , 99% percent is through Supervised Learning .
Let's take a look at Supervised Learning-
IBM describes it as follows-
It is defined by use of labeled datasets to train algorithms that classify data or predict outcomes accuarately.As input data is fed into the model , it adjusts its weights until the model has been fitted appropriately , which occurs as a part of cross validation process. Supervised Learning helps organizations solve for a variety of real-world problems at large scale such as classifying spam in a seperate folder from your inbox.
A model is an algorithm which feeds on data .In supervised learning , the training of model happens as follows -the model is provided with lots of data , for each data item it gives its own prediction then the predicted result is compared with actual result and the "cost"(error) us determined . The weights in model are adjusted such that the cost function is minimized or the model "fits" the training data in vest possible way . So , based on this training it provides best possible prediction for the test data .
It essentially mimicks the way humans learn .For example , when preparing for a exam , you feed yourself with more and more practice questions (training data ) so that you give best prediction in the exam (test data), you are the model .
The two most basic algorithms of Supervised Learning are -
1.Linear Regression
2.Logistic Regression
We will take a look at them one by one in next blog.