Difference between Classification and Regression


Classification: Classification is a method of evaluation a function which can divide the dataset classes according to different parameters. It classification, a computer program categorizes the dataset into different classes based on having raining on used to estimate whether an input value is in the pre-determined group or not.
E-mail spam detection can be considered as perfect example of classification. On the basis of millions of e-mails on different parameters, the model is trained. So, whenever a new email is received, it detects whether the email is spam or not. The spam emails are moved to the spam folder.

Regression: Regression is a method of evaluation a function which estimates continuous value based on its input variables. Instead of using classes or discrete values, continuous real values are identified. Prediction of market trend, prediction of house prices use regression model because in such models, continuous variables are predicted.
For example if we are estimating the possibility of rain using regression, a probability associated with the rain will be resulted. Here, the labels of within rain or no rain would not be shown.



Classification Regression
1. The classification method is used to evaluate a function based the associated probability will be data. 1. Regression method is used to evaluate a function for estimating continuous quantity based on its input is variables.
2. Mapping of values to predetermined classes is focused. 2. Mapping of values to continuous output is focused.
3. Calculation is done by measuring accuracy. 3. Calculation is done by measuring root mean square root.
4. Predicted data are unordered in nature. 4. Predicted data are ordered in nature.
5. Here, decision boundary is predicted that can segment the dataset into different classes. 5. Here, the best fit line is predicted. So regression can estimate the output more accurately.
6. Identification of spam emails, speech recognition, identification of cancer cells etc. problems are solved by classification algorithms. 6. Weather forecast, price of the house prediction etc. problems are sowed by regression algorithms.
7. Logistic regression k-nearest neigh bours, support vector machines, kernel SVM, naive bayes, Decision tree classification, Random forest classification are the types of classification algorithms. 7. Simple linear regression, multiple linear regression, polynomial regression, support vector regression, Decision tree regression, Random forest regression are the types of regression algorithm.




Contributed By: Romana Rahman Ema