Machine Learning and its Types


Machine Learning is the scientific study of algorithm and statistical model that the computer system used to effectively perform a specific task without explicit instruction 


Machine learning is the science and art of programming computer so they can learn from data

The process of machine learning begins with data so if you have lots of data then the machine learning model perform very well

Types of machine learning 

1) Supervised learning 

2) Unsupervised learning

3) Reinforcement learning

4) Semi-Supervised learning

Supervised Machine learning

In Supervised Machine learning model, the Algorithm learns from label dataset (which have both input and output parameters ) and  generate a reasonable prediction on new data  

Example – such as spam/not-spam email or predict the price of the house

There is 2 type of supervised learning model

–  Classification model

Classification model

–  Regression model

This image has an empty alt attribute; its file name is 1*ly54PqEsHuVvOVZf8WfeDQ.png
Regression model

In a classification problem, we have discrete output and in regression problem, we have continuous output

Example of Supervised Learning Algorithms:

  • Linear regression
  • Logistic regression 
  • K-nearest neighbors
  • Gaussian Naive Bayes
  • Support Vector Machine (SVM)
  • Decision Trees
  • Random Forest

Unsupervised Machine learning

Unsupervised learning is a type of machine learning used to draw inferences from datasets consisting of input data without labeled responses and tries to make sense of extracting features, co-occurrence and underlying patterns of data

The goal for unsupervised learning is to model the underlying structure or distribution in the data in order to get more inside from data
Example – pattern recognition, Recommender systems

Example of Unsupervised Learning Algorithms:

  • K-means clustering 
  • Anomaly detection 
  • Autoencoder 
  • Principal component analysis (PCA)
  • Association rule learning

Visualization algorithms like PCA and t-SNE are also good examples of unsupervised learning algorithms here is the visualization of the iris dataset 

Reinforcement learning 

Reinforcement learning  is a type of machine learning the learning system called the agent in this context agent can observe the environment select and perform an action in the environment and get rewards or penalties from the environment the main goal here is to get maximum rewards from the environment

A policy defines what action the agent should choose when it is in a given situation to get maximum rewards

Example – many robots implement Reinforcement Learning algorithms to learn how to walk and DeepMind’s AlphaGo program is also a good example of Reinforcement Learning

Example of Reinforcement Learning Algorithms:

  • Q-learning
  • sarsa Algorithms

Semi-supervised learning

Semi-supervised learning is of the combination of labeled and unlabeled data in other way combination of Supervised Learning and Unsupervised Learning methods.

Many real-world machine learning problems fall into semi-supervised learning Algorithm

Above image is Example of semi-supervised learning 

Most semisupervised learning algorithms are combinations of unsupervised and supervised algorithms