July 10, 2024

Alisa Loges

Global Web

Category-Specific Machine Learning Cheat Sheets

3 min read


Machine Learning is an incredibly complex topic. There are so many algorithms and concepts that it can be difficult for beginners to take the first steps. This cheat sheet is intended to give you a comprehensive introduction to machine learning with an emphasis on deep learning. It is broken down into sections which each focus on one category of machine learning algorithms: classification, clustering, regression etc. The hope is that you’ll be able to better understand each concept by reading through this guide and then going back later when you need more information about specific areas within machine learning!


  • Identify the problem. Is there a particular problem you want to solve? Do you want to classify data into different categories, or predict if an event will occur?
  • Define the problem. What exactly does “classify” mean in this context? How many different types of classes do we have? Are they all equally important, or are some more important than others?
  • Set the context: Who will use this solution (business users or developers), and what will they use it for (analyzing large datasets versus training models)?
  • State your objective: Are there any constraints on how accurate your model needs to be (e.g., acceptable false positive rate)? Can we afford more time per query over better accuracy if needed (i.e., trade-off between speed vs accuracy).


Clustering is a unsupervised learning technique that aims to group data points into clusters of similar data. Clustering can be used for finding patterns in unlabeled data, or it can be used as part of an overall process where you first use clustering to find similar groups of items, then use those groups as training examples for another machine learning algorithm (such as k-means).

Clustering techniques include:


Regression is a supervised machine learning technique that predicts continuous values. It’s used to estimate the relationship between variables, such as predicting how much money you’ll make based on your age and gender (though this example isn’t exactly helpful).

Regression is also useful when you have a large dataset, but don’t know what questions to ask it yet. You can use regression as an exploratory tool by asking questions like “What are all the different factors that affect my income?” Or, if you have some idea of what might influence your income–like age or gender–you could ask: “How do these particular factors affect my income?”

Text Analytics

Text analytics is the process of deriving meaningful information from text. Text analytics can be used to answer questions about the content of a document, such as “what topics are discussed in this text?” or “what is the sentiment of this document?”.

Text analytics is also used for classification and clustering tasks, where you want to group documents together based on their common characteristics. For example, one might want to group all emails containing offers for credit cards into one category (because these emails will likely contain phishing scams).

This cheatsheet is a comprehensive introduction to machine learning with an emphasis on deep learning.

Machine learning is a branch of artificial intelligence that enables computers to learn from data. It’s used in many areas, including robotics and healthcare–and it can be applied to your business.

Machine Learning (ML) is an umbrella term for a set of techniques that allow computers to learn without being explicitly programmed. Instead, the machine uses algorithms (sets of rules) to analyze data and make predictions based on what it has learned from previous experience or observations.

The goal here isn’t just for you to understand how ML works but also how you might use it as part of your business strategy


There are many different types of machine learning, but they all have one thing in common: they use algorithms to “learn” from data. This means that a computer program can be trained on a set of input data and then make predictions about new data based on what it learned from previous examples. In this way, machine learning systems are able to automatically improve their performance over time as they receive more information about their environment or task at hand.