Different types of Machine Learning - Part 02

Hello and Welcome back! In the previous article, we discussed the basics of machine learning and its importance. Today, we will explore the different types of ML, already you have seen the different types of ML in the last article also and now we will discuss them in detail.

Machine learning algorithms can be broadly divided into 4 major types,

1. Supervised ML
2. Unsupervised ML
3. Semi-Supervised ML
4. Reinforcement learning

Let’s discuss one by one in detail.

1. Supervised ML:

Supervised learning is a type of machine learning, where the algorithm is trained on a labeled dataset. In supervised learning, the input data and their corresponding output values are provided to the algorithm, and the algorithm learns to map the input data to the correct output value.

Supervised learning can be used for a variety of tasks, such as classification and regression. In classification, the algorithm is trained to predict a discrete output, such as whether an image contains a cat or a dog. In regression, the algorithm is trained to predict a continuous output, such as the price of a house based on its features.

You can simply remember this “If the dataset has both input and output data, then it is Supervised ML”.

Note: Discrete data is a numerical type of data that includes whole, concrete numbers with specific and fixed data values determined by counting. Continuous data includes complex numbers and varying data values measured over a particular time interval.

Types of Supervised ML

Again Supervised ML is divided into two types:

1. Regression.
2. Classification.

1] Regression:

Regression is a type of supervised machine learning algorithm used for predicting numerical values. It involves finding the relationship between a dependent variable and one or more independent variables.

In regression, the goal is to build a model that can predict the value of the dependent variable based on the values of the independent variables. The independent variables are also known as predictor variables, and the dependent variable is the variable being predicted.

Or it’s simple, If the output column is a numerical value column or if you want to predict a numeric value, then it is a regression.

Where do we use regression?

Regression is commonly used in fields such as finance, economics, healthcare, and many more. For example, a regression model can be used to predict the sales of a product based on the amount spent on advertising or to predict the risk of heart disease based on factors such as age, gender, and blood pressure.

2] Classification:

Classification is a type of supervised machine learning technique where the goal is to predict the categorical class label of a new instance based on past observations. In other words, it is the process of identifying to which of a set of categories a new observation belongs, based on training examples of known categories.

Example: When you have a Yes or No kind of output column, you can use this data. consider the problem of classifying emails as spam or not spam. We can train a classification model on a dataset of labeled emails, where each email is labeled as spam or not spam. The model then learns patterns and relationships between the features (e.g. words, phrases, sender, etc.) and the corresponding labels. When presented with a new, unseen email, the model will predict whether it is spam or not spam based on these learned patterns.

2. Unsupervised ML:

Unsupervised ML is a type of ML that involves training models to identify patterns and relationships within data without explicit guidance or labels. In other words, instead of being given labeled data to learn from, unsupervised ML algorithms analyze and identify patterns in unlabeled data on their own.

In simple, Input is given and output is not given. Then we use unsupervised machine learning.

What is the goal of this type?

The goal of unsupervised ML is to uncover previously unknown structures or relationships within the data, such as clusters, groups, or anomalies. This type of ML is particularly useful when dealing with large amounts of unstructured data, such as images, text, or sensor data.

Where do we use this type?

Unsupervised ML algorithms are commonly used in a wide range of applications, including clustering, anomaly detection, and dimensionality reduction. By identifying patterns and relationships in unlabeled data, unsupervised ML can help businesses and organizations gain new insights, improve decision-making, and streamline processes.

Types of Unsupervised ML

Again we have 4 different types of Unsupervised ML:

1. Clustering.
2. Dimensionality Reduction
3. Anomaly Detection.
4. Associative Rule Learning.

1] Clustering:

Clustering is a technique in which data points are grouped based on their similarity. The goal is to partition the data into groups, or clusters so that the data points within a cluster are more similar to each other than to those in other clusters.

You can see in the below image, how data is clustered or grouped.

2] Dimensionality Reduction:

Dimensionality Reduction is a technique used to reduce the number of features or variables in a dataset. The goal of dimensionality reduction is to simplify the dataset while retaining as much information as possible.

In many cases, datasets can have a large number of features that are irrelevant, redundant, or noisy, which can make it difficult to analyze the data or build models. Dimensionality reduction can help to eliminate these features and reduce the complexity of the dataset, making it easier to work with and allowing for more accurate analysis and modeling.

There are two main types of dimensionality reduction techniques:

1. Feature Selection: In feature selection, a subset of the original features is selected based on some criterion. This technique discards the irrelevant and redundant features from the dataset.
2. Feature Extraction: In feature extraction, a new set of features is created that summarizes the information contained in the original features. This technique transforms the dataset into a lower-dimensional space where the most relevant and informative features are retained.

3] Anomaly Detection:

Anomaly detection is a common task in unsupervised ML, which involves identifying data points that are significantly different from the majority of the data. Anomaly detection can be used for a wide range of applications, such as fraud detection, network intrusion detection, and equipment failure prediction.

In short, finding the outliers.

4] Associative Rule Learning:

Associative Rule Learning is a technique in unsupervised ML that identifies interesting relationships or associations between variables in large datasets. It is also known as Association Rule Mining or Market Basket Analysis. The goal of associative rule learning is to discover interesting patterns, correlations, and relationships within the data without any prior knowledge or assumptions about the variables.

In this technique, the algorithm searches for frequent item sets, which are groups of items that frequently occur together in the dataset. These itemsets are then used to generate rules that describe the relationships between the variables. The rules are expressed in the form of “if-then” statements, where the antecedent (the “if” part) represents the condition or criteria, and the consequent (the “then” part) represents the outcome or result.

For example, a rule generated by the algorithm might be “If a customer buys bread and milk, then they are likely to buy eggs.” This rule can be used to make recommendations to customers, such as suggesting that they purchase eggs when they buy bread and milk.

3. Semi-Supervised

Semi-supervised learning is a type of machine learning where the algorithm is trained using both labeled and unlabeled data. In traditional supervised learning, all of the data is labeled, meaning that each data point has a corresponding output or target value. However, in semi-supervised learning, only a subset of the data is labeled, and the algorithm must use the information from both the labeled and unlabeled data to make predictions.

Some common applications of semi-supervised learning include natural language processing, computer vision, and speech recognition. In these domains, it is often difficult or impractical to label large amounts of data, but there is still a large amount of unlabeled data available. Semi-supervised learning can help to make use of this unlabeled data and improve the performance of machine learning models.

4. Reinforcement learning

Reinforcement learning is a type of machine learning that focuses on training an agent to make decisions based on its environment. In RL, an agent interacts with an environment and receives feedback in the form of rewards or punishments based on its actions. The goal of the agent is to learn a policy that maximizes its cumulative reward over time.

Reinforcement learning is often used in scenarios where there is no labeled data available, and the agent must learn from trial and error. Examples of RL applications include game-playing agents, robotics, and autonomous vehicles. RL is also used in recommendation systems and online advertising to optimize user engagement and conversion rates.

We have explored four major types of machine learning, including supervised learning, unsupervised learning, semi-supervised learning, and reinforcement learning. Each of these types has its own unique strengths and limitations, and practical applications in various domains. By understanding these types of machine learning, we can choose the right algorithm for the task at hand and achieve more accurate and efficient results. As machine learning continues to advance, it opens up new opportunities for innovation and discovery across various industries. So, let’s keep exploring and learning together, as there is still so much to discover in the exciting world of machine learning!

…………................………TO BE CONTINUED………………………………..

Thank you for reading today's article on types of ML. I hope it has provided you with a better understanding of the various types of ML and their applications. I encourage you to further research these types to enhance your knowledge.

Moving forward, in our next article, we will delve into the challenges faced in the field of ML and the steps involved in solving them. This will enable us to gain a better understanding of the process involved in solving ML problems. Thank you for your continued interest in our articles.

Previous article: 1. Introduction to Machine Learning

Next article: 3. Challenges and steps involved in solving machine learning problems.

Keep learning…….??

ALL LOVE NO HATE