All the Basics You Need to Know About Machine Learning

Photo by Andy Kelly on Unsplash

Machine learning! A big buzz word in the industry and academia as well. Before you dive into machine learning basics you need to know what Artificial intelligence is? Some think these two are the same but it is not.

Artificial intelligence (AI) is a technique that enables machines to imitate human behavior. For example, Apple Siri and self-driving cars like tesla can be identified. And Machine learning (ML) is a subset of AI. And machine learning (which uses statistical methods) focuses mainly on the designing of the systems thereby allowing them to learn and make predictions based on some experience which is the data in most cases of machines. In other words, using machine learning, machines can make data-driven decisions.

Snapchat’s filters which use augmented reality and ML and how Netflix identifies what are the movies that you like to watch next automatically are some of the applications of machine learning.

Deep learning which is another buzzword in this field is a particular kind of ML that is inspirited by the functionality of the brain cells called neurons which led to the concept of artificial neural networks.

AI vs ML vs DL

There are 3 main areas in Machine learning;

1. Supervised learning
2. Unsupervised learning
3. Reinforcement learning

Let’s dig deep into these categories. 🔍

Supervised learning

In supervised learning, we use known or labeled data for the training data. Since the data is known, the learning is, therefore, supervised. The input data goes through the Machine Learning algorithm and is used to train the model. Once the model is trained based on the known data, you can use unknown data into the model and get a new or predicted response.

Supervised learning normally uses classification and regression techniques to develop predictive models.

Classification techniques predict discrete responses. Typical classification applications are medical imaging, speech recognition, and credit scoring.

You can use classification if your data can be tagged, categorized, or separated into specific groups or classes. Some of the common algorithms for performing classification are;

1. Support vector machine (SVM),
2. K-nearest neighbor,
3. Naïve Bayes,
4. Discriminant analysis,
5. Logistic regression
6. Neural networks

Regression techniques predict continuous responses. For example, changes in temperature or fluctuations in power demand. Typical applications are electricity load forecasting and algorithmic trading.

You can use regression techniques if you are working with a data range or if the nature of your response is a real number, such as temperature or the time until failure for a piece of equipment. Common regression algorithms are;

1. Linear regression
2. Stepwise regression
3. Decision trees
4. Ensembles
5. Neural networks

Unsupervised Learning

Unsupervised learning finds hidden patterns or structures in data. It is used to draw inferences from datasets consisting of input data without labeled responses.

Clustering is the most frequently used unsupervised learning technique. It is used for exploratory data analysis to find hidden patterns or groupings in data. Applications for cluster analysis include gene sequence analysis, market research, and object recognition.

Some common algorithms for performing unsupervised learning are;

1. K-means and k-medoids,
2. Hierarchical clustering,
3. Gaussian mixture models,
4. Hidden Markov models,
5. Self-organizing maps,
6. Fuzzy c-means clustering, and
7. Subtractive clustering.
8. Apriori
9. Singular value decomposition
10. Principal component analysis

Reinforcement learning

In reinforcement learning, the algorithm discovers data through a process of trial and error and then decides what action results in higher rewards. There are three major components in reinforcement learning: the agent, the environment, and the actions. The agent is the learner or decision-maker, the environment includes everything that the agent interacts with, and the actions are what the agent does.

Reinforcement learning occurs when the agent chooses actions that maximize the expected reward over a given time. This is easiest to achieve when the agent is working within a sound policy framework.

Hopefully, I think you were able to gain some new knowledge by going through this article. Thank you! 😃 🎉

References:

What is Machine Learning and How Does It Work?

What Is Machine Learning? | How It Works, Techniques & Applications