What is Machine Learning ?

Machine Learning (ML) is the subfield of Artificial Intelligence . ML models uses the past experience or knowledge gained from the input data to perform specific tasks namely classification, regression and clustering without manual intervention or monitoring.

What are different types of Machine Learning algorithms ?

Supervised Learning : The model built through learning procured from the labelled data , which means the data consists of features and label for each instance . Examples : Classification , Regression
Unsupervised Learning : The model built through learning procured from the unlabelled data, which means the data consists consists of only features for each instance and there is no label associated with it. Examples : Clustering.

List the supervised machine learning algorithms used for classification.

Naive Bayes
K Nearest Neighbor
Decision Tree
Neural Networks
Support Vector Machine Classifier

List the supervised machine learning models used for regression.

Linear Regression
Support Vector Machine Regressor
Neural Networks
Decision Tree Regressor

List the unsupervised machine learning models .

K means clustering
Principal Component Analysis
Hierarchical clustering
Fuzzy based clustering

What is the structure of Neural Network ?

Neural Network (NN) consists of neurons, which is the basic computational unit. NN (Multilayer perception) has three layers namely input layer , hidden layer and output layer.

Why is an activation function used in hidden layer ?

Activation function is used in order to introduce non-linearity to solve the high computational problems in a more simpler manner.
Types of Activation Functions are : Sigmoid , Tanh, ReLU and Softmax.

When do we use Regression models?

The most important condition to use Regression model is when the label of the data is of continuous type.

Example : House Price Prediction, Stock value Prediction.

When do we use Classification models ?

The most important condition to use classification model is when the label of the data is of discrete value type.

Example : Cancer Prediction.

What is entropy in decision tree?

Entropy is the measure of impurity present in the particular node of the decision tree.