ML algorithms has truly learned the relationship between input and output. To do that, we divide the data into two parts – training data and test data. We use training data to learn the given model. And then, we use the test data to evaluate the model. So, What learning a model actually means is that we will study it in other concepts down the line.
While there are different classes of ML algorithms, they differ in the way the input is presented to the ML algorithms and the output that is expected.
Classes Of ML Algorithms
1. Supervised Machine Learning :-
for Supervised ML, The Input Is Tagged With Expected Outputs And We Try To Predict The Outputs For Future Inputs Too. Some Examples Of Supervised Learning Are As Follows:
Various types of supervised learning:
- Classification
- Regression
The main difference between classification and regression is the form of the output variable. Whereas, If the output variable is discrete, it is classification, and if it is continuous, it is regression.
You can think of discrete data as something that is counted, and continuous data as something that is measured. Discrete data refers to things like movie genres, categories of mail (spam/not spam) and so on. Continuous data refers to data like the height of a person, stock price, price of a house, and so on.
Classification
It also involves finding a set of categories a new data point it belongs to. The output value is a class or a category.
Examples:
1)Deciding if the image – a cat or dog.
2)Deciding if an email is spam or not.
3)Predicting whether a patient has cancer or not.
4)Predicting whether the value of a stock will rise or drop.
Regression:-
It also involves predicting a continuous value, given a set of inputs. The output value is continuous.
Examples:
1)Predicting the stock price given different factors.
2)Predicting the sales of a company for a given year.
3)Also, Predicting runs/goals scored by a sports team.
4)Assigning credit rating.
Consider a situation where we have the data from houses in the city.
- So, Predicting the price of the house, given various factors like the area of the house, location, no of bedrooms, etc: is Regression.
- Predicting if a house will get sold in the next few months, given various factors like the area of the house, location, etc. is Classification.
2) Unsupervised Machine Learning:-
Unsupervised Machine Learning is the task of inferring a function to describe hidden structure from “unlabeled” data (a classification or categorization is not included in the observations).Unsupervised learning is where you only have input data (X) and no corresponding output variables.The goal for unsupervised learning is to model the underlying structure or distribution in the data in order to learn more about the data.So, in this case, we try to find homogeneous groups between data which we refer to as clusters. However, objects in a cluster exhibit similar behaviour.
Clustering:
1) Unsupervised Learning Technique.
2) Grouping objects in clusters.
Examples:
1)Segmenting customers into various groups.
2)Also, Online shopping recommender system: Grouping things bought together.
3)Grouping Similar Animal Photos.
3) Reinforcement Learning:-
Used to solve interacting problems where the data observed up to time “t” is considered to decide which action to take at the time “t+1”. Also used for AI, when training machines to perform tasks such as Walking, etc.The agent learns from the environment by interacting with it and receiving reward/punishment.
Imagine a child walking towards a fireplace. They will feel warm and go closer towards the fire. This is like a reward. When the child gets too close to the fireplace, they will feel hot and step back. This is punishment. Based on reward and punishment, the agent learns to navigate. An agent playing the Mario game is an example of reinforcement learning.
Written By: THOMAS VENGAZHIYIL ALEX
Reviewed By: Viswanadh
If you are Interested In Machine Learning You Can Check Machine Learning Internship Program
Also Check Other Technical And Non Technical Internship Programs
