Hi folks, In the field of machine learning we all know the type of problems are different, sometimes we predict the value on previous set of data – Where data learn from available dataset, Or sometimes grouping them into some cluster. So today we are going to see what these terms are – **Clustering, Classification and Regression** means in Data science field. let’s dive into this concept.

Generally machine learning algorithms are categorised on the basis of output type and type of problem that need to be addressed. So these algorithm are divided into three categories –

- Classification
- Regression
- Clustering

In above example Classification and Regression are the example of Supervised algorithm where Clustering is unsupervised algorithm.

When the output variables are continuous then it is a Regression problem whereas when it contain discrete value it is Classification problem. Clustering algorithms are generally used when we need to create cluster based on the characteristics of data points. Let’s see above points in details.

**Classification** : –

Classification is the type of supervised machine learning, For any given input, the classification algorithm help in the prediction of the class of the output variables. there can be multiple type of classification are – binary classification, multi-class classification.

**Types of classification – **

- K – Nearest Neighbour
- Logistic regression
- Decision tree
- Random forest
- Naive Bayes
- SVM (Support vector machine)

**Regression** : –

Regression is the type of supervised machine learning, When the output is continuous like age, height etc. one of very popular regression algorithm is **Linear Regression.**

**Types of Regression – **

- Linear Regression
- Ridge Regression
- Lasso

**Clustering** : – Clustering is unsupervised machine learning algorithm, it is used to group data point having similar characteristics as cluster.

Clustering is divided into two groups

- Hard clustering – In hard clustering, the data point is assigned to one of the clusters only.
- Soft clustering – It provides a probability likelihood of a data point to be in each of the clusters.

**Types of clustering** –

- K-Means
- DBSCAN
- Mean-shift algorithm
- Affinity propagation

So above all information discussed in brief, will go one by one in detail. Hope basic Understanding about classification, regression and clustering are clear now.

Thanks.

## Leave a Reply