Classification in Machine Learning | Easy Explanation
Updated: Aug 10
There are basically four types of techniques in machine learning. Those are Classification, Regression, Recommendation, Clustering. We have to choose any techniques according to our problem. Here, you are going to give your valuable time to classification in machine learning. So, without wasting any time lets start the journey.
What you will learn?
What is classification in machine learning?
Example of classification in machine learning
Actions performed by classifier systems.
Applications of classification in machine learning.
Classification algorithms in machine learning.
What is classification in Machine Learning?
classification in machine learning is a technique that determines how new data should be classified into existing categories with the help of a given dataset those categories.
It performs on both structured as well as unstructured data. Classification predictive modeling is the task for approximating a mapping function (f) from input variables (X) to discrete output variables (y).
It is also called the Categorization technique. It is a Supervised Learning.
Example of classification in Machine Learning
Suppose there is a credit card company that receives thousands of applications for issuing a new credit card. Applications contain much information like annual salary, sex, age, location, credit record, etc.
Here, the task of the classification algorithm is to classify the applicants into categories like who have a bad credit record, good credit record, and mixed credit record.
Imagine in a hospital, the emergency room has many features like a heart condition, blood pressure, age, etc. To analyze before deciding whether a patient has to be put into an emergency as it is a costly proposition and patient who can afford them and survive are given the top priority. The problem here is based on the available features, classify the patients into high or low-risk patients.
Actions performed by system classifier system.
A new data model is initially prepared with the help of any classification algorithms
prepared model is tested
Later, the prepared model is used to examine the class of new data.
Applications of classification in Machine Learning.
Weather prediction- predicting whether on a certain day it will rain or not.
Share prediction- prediction of the company share price will fall or not.
Detection of credit card Fraud- based on employing historical records of previous frauds, here classifiers predict whether future transactions may turn into fraud or not.
E-mail spam- depending on the characteristics of previous emails it predicts whether a newly received email is spam or not.
Classifications algorithms in machine learning
Logistic regression- It is a classification algorithm in machine learning which uses one or more than one independent variable to determine the outcome.
Naive Bayes-It predicts that the presence of a particular feature in a class is individual to the presence of any other feature.
Stochastic Gradient Descent-It is useful when the sample data is a very large number. It also supports different loss functions, penalties for classification.
K-Nearest Neighbours- It is a supervised learning algorithm that stores all cases and then classifies new cases with the majority vote of its K neighbors.
Decision tree- It is a mostly used supervised learning algorithm for classification problems.
It uses a tree-like model to make decisions.
Random Forest- It is a popular ensemble supervised learning algorithm. Ensemble means it takes weak learners and makes them work together to form one strong predictor.
These are some basic definitions related to the respective algorithm. To know more
Click on names of the algorithm mentioned above. "Links of famous pages "