Overview about Machine Learning

Machine learning is a type of artificial intelligence (AI) that provides computers with the ability to learn without being explicitly programed. Machine learning focuses on the development of computer programs that can change when exposed to new data.

By this image, you can see the AI development until now.

• AI: Ability of machine to imitale intelligent human behavior.
• Machine Learning: Application of AI that allows a system to automatically learn & improve from experience.
• Deep Learning: Application of Machine Learning that uses complex algorithms & deep neural nest to train a model.

What is “learning” in ML?

“A computer program is said to learn from experience E with respect to some tasks T and performance measure P, if its performance at tasks in T, as measured by P, improves with experience E.”

1. Some basic problems that need to be solved in ML

1.1. Classification

The data is categorized under different labels according to some parameters and then the labels are predicted for the data. It needs to learn decision making, choosing one decision among “N” decisions by looking at past experience.

Eg. Classifying emails as either spam or not spam.

1.2. Regression

It predicts valued output. It needs to learn a predicting particular outcome in a context by looking at past experience.

Eg. Predicting a person’s income from their age, education.

1.3. Clustering

Clustering is the task of partitioning the dataset into groups, called clusters. The goal is to split up the data in such a way that points within single cluster are very similar and points in different clusters are different.

Eg. Clustering customers based on buying behavior that based on the tracking on an e-commerce site.

2. Grouping machine learning algorithms

2.1. Supervised learning

Supervised learning uses a training set to teach models to yield the desired output. This is the most common group of algorithms in ML. Classification and Regression algorithms are two examples in this group.

2.2. Unsupervised learning

Unsupervised learning uses machine learning algorithms to analyze and cluster unlabeled datasets. These algorithms discover hidden patterns or data groupings without the need for human intervention. Clustering is an example in this group.

2.3. Semi-supervised learning

Semi-supervised learning is a learning problem that involves a small number of labeled examples and a large number of unlabeled examples. You train an initial model on a few labeled samples and then iteratively apply it to the greater number of unlabeled data.

2.4. Reinforcement learning

Reinforcement learning may not require the training dataset, the model learns to make decisions by communicating with the environment. The decisions are make continuously and receive feedback from the environment to reinforce behavior.