Types of Machine Learning

 Types Of Machine Learning

Machine learning is a type of artificial intelligence that allows software applications to become more accurate at predicting outcomes without being explicitly programmed to do so.

In simpler way, machine learning allows the user to feed the computer algorithm an immense amount of data and have the computer analyse themselves and make predictive and accurate decision based on only the input data.

Based on the method and the way of learning there are four types:

1.Supervised Machine Learning

2.UnSupervised Machine Learning

3.Semi-Supervised Machine Learning

4. Reinforcement Learning

1. Supervised Machine Learning:

Supervised machine learning, as the name indicates, has the presence of supervisor as a teacher. Means, supervised learning is when we teach or train the machine using data that is well labelled.

Here labelled means some of the input data is already mapped to the output.

Suppose, you are given a basket filled with different kind of fruits such as apple and banana.

If the shape of object is round and is red in color then it will be labelled as - Apple

Supervised learning is classified into two categorial of algorithms.

• Classification: is used to solve the classification problems in which the output variable is categorical such as 'yes' or 'no', 'disease' or 'no disease'.

Some popular classification algorithm are: 

1.Random Forest algorithm

2.Decision Tree algorithm

3.Logistic Regression algorithm

4.Support vector machine learning algorithm

• Regression: is used to solve the regression problems in which there is a linear relationship between input and output variables.

Some popular regression algorithm are:

1.Simple linear regression algorithm

2.Regression trees

3.Non Linear regression algorithm

4.Bayesian linear regression

5.Polynomial regression

2.Unsupervised Machine Learning

Unsupervised learning, as the name suggests there is no need for supervision, it means the machine is trained using the unlabelled dataset, and the machine predict the output without any supervision.

The main goal is unsupervised learning is to group or categories the unsorted dataset according to the similarities, patterns and differences.

For instance, suppose it is given an image having both cats and dogs that has never seen before.

Thus, the machine has no idea how to categorise them, but it can categorise according to the similarities, patterns and differences.

Unsupervised learning is classified into two categorial of algorithms.

• Clustering: is used when we want to find inherent groups from the data. 

       It is a way to categorise such as if there is one side there are more similarities and on the other side having fewer and no similarities.

       

       An example is that grouping the customers by their purchasing behaviour

• Association: is learning problem where you want to discover rules that describe large portions of your data, such as people that buy x also tends to buy y.

Some popular algorithms are:

1.K-means clustering

2.KNN

3. Apriori algorithm

4.Anomaly detection

5.Eclate

6.FP-growth algorithm

3.Semi-Supervised machine learning algorithm

Semi-supervised learning is a type of Machine Learning algorithm that lies between supervised and unsupervised machine learning and operates on the data that consist of a few labels, it mostly consist of unlabelled data.

The main aim of semi-supervised learning algorithm is to effectively use all the available data.

Practical application of semi-supervised learning is Speech analysis,Internet content classification.

4.Reinforcement learning

Reinforcement learning works on a feedback-based performance, in which an AI agent automatically explore its surrounding by hitting&trail, taking actions, learning from experiences and improving its performance.

It is based on rewarding desired behaviour and punishing undesired ones.In short, it learn through trial and error and improve its performance.

Suppose, there is an AI agent within maze environment, and his goal to find diamond.The agent interacts the environment by performing some actions, and based on the actions, the state of the agent gets changed  and it receives rewards and penalty as a feedback.

The agent continue doing these steps and by doing these actions, he learns and explores the environment.

Real-world use cases of reinforcement learning

1.Video Games

2.Robotics

3.Text mining

4.Resource management

Thanks for Reading and Happy Coding💜