BRIEF VIEW ON MACHINE LEARNING:

on

WHAT IS MACHINE LEARING?


Machine learning is the most important subset of artificial intelligence. Machine learning is a branch of AI. Machine learning trains the machine to do so. on the idea that systems can learn from data, identify patterns and make decisions with minimal human intervention.

                   HERE ARE 4 METHODS OF MACHINE LEARNING. 

  1.                                   
                                     1.Unsupervised learning. 
  1.                                 2.Supervised learning. 
  1.                                 3. Semi-supervised learning. 
  •                                 4.Reinforcement learning. 

 HOW TO USE THESE FOUR ALGORITHMS?

                   
  • If you want to perform dimension reduction .Then use principle components analysis.
  • if you need a numerical quickly .Use decision tree or logistic regression.
  • if you need a hierarchical result, Use hierarchical clustering.
some times more then one branch will apply. The only way to find the very best algorithm is to try all of them.

Unsupervised learning.

  when performing unsupervised learning the machine is present with totally unlabled data. It is asked to discover the intrinsic patterns that underlies the data . such as a clustering structure.A low dimensional manifold,or a spare tree and graph.

clustering:
     grouping a set of dataset into several groups.

Dimension reduction:

reduction the number of variables under consideration,redundant or irrelevant to the task.


supervised learning.

supervised learning algorithms make predictions based on a set of example.

for example:
                      Historical sales can be used to estimate the future price.


stages involved:
  • Classification.
  • Regression.
  • Force casting.
classification: 
                       The data xare being used to predict a categorical variable.
   for example:
                       In this case when assigning a lable or indicator. either dog or cat to an image.
  •  2-Lables are called binary classification.
  • More than 2-Lables are called Multi-class classification.
  • Supervised learning are called as classification learning.


Regression: 
                  when predicting continuous values the problem became a regression problem.
                    
Forecasting:
                      This is the process of making predictions about the future. based on the past and present                       data.         

for example:
                    A common example might be estimation of the next year sales based on the sales of the current year and previous years.

semi-supervised learning:

                                      The challenging with supervised learning is that labling data  can be expansive and time consuming . small amount of labled data to improve the learning accuracy. 

Reinforcement learning:

                                    Reinforcement learning analysis and optimizes the behaviour of an agent based on the feedback from the environment. trail-and-error and delayed reward distinguishing reinforcement learning from other techniques.
                 








Comments

Post a Comment