# What Type of Data Is Used in Machine Learning?

//

Larry Thompson

Machine learning is a fascinating field that involves teaching computers to learn from data and make predictions or decisions without being explicitly programmed. Central to the success of machine learning algorithms is the use of data.

But what type of data is used in machine learning? In this article, we will explore the different types of data that are commonly used in machine learning, including numerical, categorical, and text data.

## Numerical Data

Numerical data is perhaps the most common type of data used in machine learning. It consists of numbers and can be further categorized into two types: continuous and discrete.

### Continuous Numerical Data

Continuous numerical data is characterized by a range of values that can take on any real number within a specified interval. Examples include measurements like temperature, weight, or height. In machine learning, continuous numerical data is often represented as floating-point numbers and used for regression tasks.

### Discrete Numerical Data

Unlike continuous numerical data, discrete numerical data can only take certain fixed values within a defined range. Examples include counts or quantities such as the number of items sold or the number of people in a group. Discrete numerical data is commonly used in classification tasks.

## Categorical Data

Categorical data represents qualities or characteristics that do not have an inherent order. It can be further divided into nominal and ordinal categories.

### Nominal Categorical Data

Nominal categorical data consists of categories with no specific order or ranking. Examples include colors, gender, or types of objects. In machine learning, nominal categorical data is often encoded using one-hot encoding to represent each category as a binary feature.

### Ordinal Categorical Data

Ordinal categorical data also represents categories, but they have a specific order or ranking. Examples include ratings, levels of satisfaction, or educational qualifications. In machine learning, ordinal categorical data can be encoded as numerical values to preserve the order and capture the relative differences between categories.

## Text Data

Text data is another type of data used in machine learning. It involves analyzing and extracting information from textual content such as articles, tweets, or customer reviews. However, machine learning algorithms cannot directly process text data – they require preprocessing steps to convert text into numerical representations.

### Bag-of-Words Representation

A common approach to representing text data is the bag-of-words model. This model treats each document as a collection of words and represents it as a vector where each element corresponds to the frequency or presence of a particular word.

### Word Embeddings

Word embeddings are another powerful technique for representing text data. They capture the semantic meaning of words by mapping them to dense vectors in a high-dimensional space. Word embeddings enable machine learning algorithms to understand the context and meaning behind words.

## In Conclusion

In machine learning, different types of data are used to train models and make predictions. Numerical data can be continuous or discrete, while categorical data can be nominal or ordinal. Text data requires preprocessing steps like bag-of-words representation or word embeddings to enable machine learning algorithms to process it effectively.

To become proficient in machine learning, it’s essential to understand these different types of data and how they can be used effectively in various tasks. So start exploring these types of data and experiment with different techniques for preprocessing and representing them!