Categories of Supervised Learning

- Overview

Supervised machine learning (ML) is a type of ML where algorithms learn from labeled data. The algorithm then uses the data to determine which label to give new data. 

Supervised learning can be divided into two categories: classification and regression.

Supervised ML models are often used to predict outcomes for unseen data. For example, they can be used to predict house price fluctuations or understand the sentiment of a message.

Some examples of supervised ML algorithms include: linear regression, decision tree, K nearest neighbors, random forest, naive Bayes.

Supervised learning is used in image and speech recognition, recommendation systems, and fraud detection.

Supervised learning is the most common type of machine learning (ML) task. It's used because it's easy to implement and deals with simple tasks. Supervised learning uses patterns to predict values in unlabeled data. It's often used in automation, when there are too many data inputs for humans to process effectively, or over large amounts of data records.

- Supervised Classification Algorithms

Classification algorithms are a type of supervised learning technique that uses a given dataset to categorize new observations into groups. 

There are two types of classification algorithms: 

• Binary classifier: A classification problem with only two possible outcomes
• Multi-class classifier: A classification problem with more than two possible outcomes

Here are some examples of supervised classification algorithms: 

• Naive Bayes: Calculates the probability that a data point belongs to a certain category. It can be used in text analysis to categorize words or phrases as belonging to a preset category.
• Random forest: A flexible algorithm that can be used for both classification and regression. It combines a collection of uncorrelated decision trees to reduce variance and create more accurate data predictions.
• K-nearest neighbors (KNN): A non-parametric classifier that uses proximity to make classifications or predictions about the grouping of an individual data point.

[More to come ...]