Components of machine learning

The three main components of machine learning include :

• Machine Learning Model

• Model training algorithm

• Model Inference Algorithm

Machine learning model

This is a file that has been trained to recognize a certain type of pattern on a dataset and providing it an algorithm that it can use to solve different problems.
The model can further be broken down into two subtypes: Supervised and Unsupervised Learning .
Machines are trained using labelled data therefore the datasets contain both input and output parameters.

Types of supervised learning

1. Continuous label (Regression)
   This technique consists of mathematical methods which help in finding the correlation between variables and enables us to predict the continuous output variable based on one or more predictor variables.

2. Categorical label (Classification)
   This technique allows a program to learn from the given dataset or observations and then classify new observations into a number of groups.

other types include :

1. Naive Bayesian Model
2. Random Forest Model
3. Neural Networks
4. Support Vector Machines

Unsupervised Learning

This technique has input data but no corresponding outputs. The goal is to find the underlying structure of a dataset and group it according to similarities, then represent that dataset in a compressed format.
Types

1. Clustering This is the process of grouping similar data together. The aim is to find similarities in the data point and group similar data points together.
2. Association This technique checks for the dependency of one item on another data item and maps accordingly so that it can be more profitable.

Model training algorithm

the algorithm is provided with training data which must contain the correct answer. it finds patterns in the training data that map the input data algorithm to the target and it outputs an ML model that captures these patterns.

Model Inference Algorithm

Using a trained model to generate predictions.