Machine Learning Models

Machine learning is an application of artificial intelligence (AI) that provides systems the ability to automatically learn and improve from experience without being explicitly programmed. Machine learning focuses on the development of computer programs that can access data and use it learn for themselves.

Now when we say, “without being explicitly programmed”, there has to be a system which does the job. Such a system is algorithm. An algorithm is a process or set of rules to be followed in calculations or other problem-solving operations, especially by a computer.

Machine learning algorithms derive their strength from their ability to learn from the available data.

We have different types of Machine Learning methods and they are based on various algorithms.

labelled and unlabeled data:

When you know the right answer to a question related to the data, then it is labelled data.

For example, you have an image, and the question is what object is in the image. If you know that the image represents a car, then it is a labelled image.

When you don’t know the right answer to the question, it is an unlabeled data.

For example, you have a lot of images and you don’t know the answer to which images belong to what category, then those are unlabeled data (images).

Types of machine Learning:

• Supervised Learning

-Classification

-Regression

• Unsupervised Learning

-Dimensionality Reduction

-Clustering

• Semi-supervised Learning
• Reinforcement Learning

1. Supervised Learning:

In supervised learning, the machine is given a dataset (i.e., a set of data points), along with the right answers to a question corresponding to the data points. Supervised learning model uses training data to learn a link between the input and the outputs.

The learning algorithm is provided with a huge set of data points with answers, i.e., a labelled dataset.

The algorithm has to learn the key characteristics within each data point in the dataset to determine the answer.

So, next time a new data point is provided to the algorithm/machine, based on the key characteristics, the algorithm should be able to predict the outcome/right answer.

1.1 Classification Problem

This refers to the ability to classify something into a distinct set of classes or

categories.

1.2 Regression Problem

Regression refers to the ability to predict values of a continuous variable

2. Unsupervised learning:

In unsupervised learning, the machine is provided with a set of data and is not provided with any right answer. Given the huge amount of data, the machine may identify trends of similarity. The algorithm will identify clusters or groups of similar items or similarity of new item with existing group.

2.1 Dimensionality Reduction

Dimensionality reduction is the process of reducing the number of random variables under consideration, by obtaining a set of principal variables. It can be divided into feature selection and feature extraction.

2.2 Clustering

we can use clustering analysis to gain some valuable insights from our data by seeing what groups the data points fall into when we apply a clustering algorithm.

3. Semi-Supervised Learning:

Semi-supervised learning falls somewhere between supervised and unsupervised learning.

Here, the machine is given a large dataset, in which only a few data points are labelled. The algorithm will use clustering techniques (unsupervised learning) to identify groups within the given dataset and use the few labelled data points within each group to provide labels to other data points in the same cluster/group.

4. Reinforcement Learning

Learning has to result in the machine sensing the external environment and choose an action based on its own state and the external environment, with the aim of maximizing a specific predefined goal.

Changing situations: for example, driving, game of chess