Machine Learning and its Algorithms overview.

Machine Learning

Making computers to do things that are not explicitly programmed, but by making computers to learn them.

In our everyday life millions of Billions of data are generated everyday, by analysing and making sensible decisions using these data we can improve the lives of people and businesses. Machine Learning employs self learning algorithms that can turn these data into useful knowledge, using which effective decision can be made. To list a few things that can be done using machine learning are identify patterns in data and make prediction about futre events based on information about past events.

There are 3 different types of machine learning in general.

We will learn about basic difference about aforementioned techniques, and learn how to implement them using Scikit

Supervised Learning

Learn a model from the existing data, it is called training data and based on that training predictions are made on unseen future data.

Lets Code Image

Supervised learning can further be divided into 2 sub categories:

Reinforcement Learning

Objective of reinforcement learning is to develop an agent which interacts with the environment via series of actions and those actions will be rewarded at every predefine state. Example: Google’s human skeleton model taught itself to walk via reinforcement learning. Lets Code Image

Unsupervised Learning

Unsupervised learning is suitable for unlabeled data or data of unknown structure. Using this technique we could extract meaningful information without guidance of known outcome variable or reward function.

Throughout this course our main focus will be building predictive classification models using supervised learning technique with Scikit learn library.

In the immediate next article, we will discuss about datasets we will be using throughout this course and how to import those datasets.