What is Machine Learning in Python? Types of Machine Learning, Steps in Machine Learning

Machine learning (ML) allows computer to learn to do things without explicit programming. Computer can learn and adapt when presented with information in appropriate way. Machine learning based on Purchase and search history.

Machine learning (ML) and artificial intelligent (AI) are both branch of computer science which are closely related, but machine learning are closely related to data mining rather than artificial intelligent (AI). Spam emails are good example of machine learning and PayPal uses machine learning to detect fraud.


There are six steps involved in Machine learning, which are;

  1. Collecting Data; This sis first step in machine learning because you cannot train your machine without data, so you need to provide some data.
  2. Cleaning Data; after collecting data you need to clean them. In this stage you have to make sure that your data have all usefully fields, so there are no other useless fields that you actuary don’t required, it can be columns or rows or anything.
  3. Analyze Data; Performing some analysis of data, it will help to train the data
  4. Train the Algorithm; It trains the algorithm based on those analysis
  5. Test the Algorithm; it will test by providing some data that actuary know what the output and they going to test whether algorithm has output which is equal to the actuary output
  6. Use it; finally you use it.



There are two types of machine learning, which are;

  1. Supervised machine learning and
  2. Unsupervised machine learning.


Supervised machine learning

Training machine learning task for every input with corresponding target, and after that machine will be able to provide target for any new input after sufficient training, known as supervised learning.

 So if you provide a machine with sufficient data it will actuary able to provide target output for any new input.  Example when you search movies in a store, the movie with large number of kisses will be romantic movie and movie with large number of kicks will be action movie.

Unsupervised machine learning

If you are training machine learning task with an only set of input not all input, after that machine will be able to find out the relationship between different inputs, it is known as unsupervised leaning.


Applications of machine learning

  1. Search engine results
  2. Voice recognition
  3. Dream reader
  4. Animated cars etc.


Can Compute see?

Mostly NO: computers can only “see” certain types of object under limited circumstances.

YES: for certain constrained problems (example face recognition).

Login to comment