Hey, readers! In this article, we will be focusing on an IMP metric of **Machine Learning – Regression v/s Classification**, in detail.

So, let us begin!!

## Let us at first discuss about Regression v/s Classification!

In the domain of Data Science and Machine Learning, Machine Learning Algorithms play a very important role in the outcome of certain predictions of a problem statement.

Machine Learning comprises two types of algorithms: **Supervised Learning and Unsupervised Learning** Algorithms.

Supervised ML algorithms work on labeled data and learn from the data provided to them as inputs in order to predict patterns from it. On the other hand, Unsupervised ML Algorithms do not learn from the historic data. Rather, these algorithms identify items having similar characteristics and group them together to form a cluster of categories.

Supervised ML algorithms have the below types of sub-divisions:

**Regression****Classification**

In regression-based algorithms, the main task of the algorithm is to perform predictions on the numeric data values and then the outcome is again a set of numeric values as a result of predictions.

Here is the list of most prominently used Regression ML algorithms:

- Linear Regression
- Decision Tree Regression Algorithm
- Random Forest Regression Algorithm
- Support Vector Regressor

On the other hand, Classification algorithms work on categorical values and thus help us predict the category of the data or the group of the data. Thus, these algorithms predicts the data values into different classes.

Few mostly used Classification ML algorithms:

- Naïve Bayes
- Support Vector Machine
- Logistic Regression
- Random Forest Classifier
- KNN

## Regression v/s Classification – Get set GO!

Having understood Regression and Classification in Machine Learning, let us now move ahead with the differences between them.

### 1. Nature of the resultant values

In Classification, the resultant set of values is categorical in nature. That is, they carry a label. For example, Gender is being classified into Male and Female.

On the other hand, Regression algorithms predict a set of values that are continuous in nature i.e. numeric data values. For example, prediction of the count of people who would opt for taxi on a sunny day.

### 2. Method of calculation

Regression algorithms calculate the best fit line so as to predict the nearest correct continuous data for the training dataset.

While the Classification algorithms detect the boundary known as decision boundary which helps it distinguish the input data values into different categories.

### 3. Type of the outcome values

The data values predicted by Regression algorithms are continuous in nature.

In Classification algorithms, the outcome is unordered in nature.

### 4. Evaluation of the model

The Accuracy of the Regression algorithms is measured using different errors such as **MAPE**, **R squared**, etc.

While the Classification Algorithms make use of Confusion Matrix, Recall, f-1 score, etc to judge the accuracy of the model

## Conclusion

By this, we have come to the end of this topic. Feel free to comment below, in case you come across any question.

For more such posts related to Machine Learning, Stay tuned and till then, Happy Learning!! 🙂