What is machine learning | what is ml | what is deep learning – complete detail

What is machine learning | what is ml | what is deep learning - complete detail

What is machine learning | what is ml | what is deep learning – complete detail

Machine learning is such an application, which gives systems such an ability that they can learn automatically and they can also improve themselves when needed.

To do this, they use their old experience only.

Machine learning technology has always focused on the development of computer programs so that it can easily access the data and later use it for its own learning.

In machine learning, learning starts from the observations of the data, for example direct experience, or instruction, finding patterns in the data and making it easier to take better decisions in fear.

The main goal of Machine Learning Technology is to invent such a technology in which computers can automatically learn that too without any human intervention or assistance so that they can easily adjust their actions accordingly.

Types of Machine Learning  – Types of Machine Learning

Its types are as follows:-

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

What is Supervised learning?

Supervised learning is a type of machine learning in which data labeled by people is used to train the machine.

In this technique, the machine uses the labeled data to create the model of the machine to understand the data sets.

Labelle data is said to be a kind of input data which is already present with the machine and by analyzing this data, the system also predicts the output data of the machine.

If I tell you in simple words, supervised learning is a process in which the system gives the correct output data to the user with the help of input data.

Supervised learning is supervised learning. For example – a student learns things under the supervision of a teacher.

Supervised learning is used in a variety of places such as Risk Assessment, Image classification, Fraud Detection, and spam filtering etc. to detect.

Types of Supervised Learning

There are mainly two types of Supervised Learning such as :-

1– Regression

Regression is a type of supervised learning. This is a technique that players use to find out the relationship between independent and dependent.

Apart from this Regression technique is also used as a method of predictive modeling in machine learning.

There are also many types of Regression such as Linear Regression, Non-Linear Regression, Bayesian Linear Regression, Polynomial Regression and Regression Trees etc.

2– Classification

Classification is an algorithm in which data is organized into different categories. Classification is used to classify the data into classes or groups.

In classification, mathematical techniques are used to classify data such as:- decision trees, linear programming, and neural networks etc.

For example – It can be used by people to divide the students of a class on the basis of their grade i.e. (average, good, excellent).

Benefits of Supervised Learning 

1- This technique greatly helps the machine to predict the output data based on the old input data.
2- In this, the user is told about the classes of objects accurately.
3 – Supervised learning model helps its users to solve real world problems like fraud detection, spam filtering etc.

Disadvantages of Supervised Learning 

1.- This learning does not enable people to do difficult tasks.
2- It takes a lot of time to predict the output data.

Unsupervised learning  – What is unsupervised learning?

Unsupervised learning Machine learning is a type that is the exact opposite of supervised learning. If we tell you in simple words, “Unlabeled data is used in this to train its machine.”

This is such a learning technique in which the machine learns things without any supervision.

Unsupervised learning is used by people to get useful insights from huge amounts of data.

Unsupervised learning model is very capable of thinking like a human being, it teaches a lot to a human being like behave, act and think like a human being etc.

Types of Unsupervised Learning

Unsupervised Learning is mainly made up of two parts:-

1- Clustering

Clustering is a method in which objects are divided into different groups, in which all the objects which are similar are kept in one group and those objects which are different are kept in another group.

Clustering also plays a big role in our normal life. For example, in a restaurant, different types of food are given to the people and cars, bikes and other vehicles are kept in the vehicle showroom.

2– Association

Association is a technique that tells people how an object is related to each other. Association is very well known for finding great relationships between variables in large databases.

Advantages of Unsupervised Learning

1- Unsupervised learning is more complex than Supervised learning technique because it does not have labeled data due to which it can easily complete complex tasks.

2- In this, it becomes much easier for the user to get the data, because it becomes much easier to get the unlabeled data in comparison to the labeled data.

Disadvantages of Unsupervised Learning

1- This learning takes a lot of time.
2- Its results are not always accurate, due to which the user does not get the correct information using it.

Semi-supervised learning – What is Semi-supervised learning technique?

Semi-supervised learning is a type of machine learning technique that is made up of both supervised learning and unsupervised learning.

It uses a small amount of labeled data to teach the machine.
In this, a large amount of unlabeled data is used to teach the machine.

Benefits of Semi-supervised learning

1- Semi-supervised learning is very easy for any user to understand.
2- Its working capacity is also high.
3- Its efficiency is also high.

Semi-supervised learning Disadvantages

1- In this technique, the user does not get to see the perfect result.
2- Its results are not always stable.

Reinforcement Learning – What is Reinforcement Learning Technique?

Reinforcement learning technique is such a learning technique in which an agent is given a great reward for doing the right thing and he is given a penalty for doing the wrong thing.

This technology is based on feedback.
In this technique, on the basis of feedback, the agent learns automatically and he improves himself.
For example- There is a robot which learns to move its hands on its own.

Benefits of reinforcement learning

1- This technique is used to achieve such results which are very difficult to achieve.

2- This technique provides accurate results.

Disadvantages of reinforcement learning

1- This technique cannot be used to solve simple problems.

2- In this technique people need more data.



Please enter your comment!
Please enter your name here