Machine Learning

Machine learning is the ability for an application to be able to learn and recognize new patterns and thus, mimicking human behaviour more closely through the use of algorithms and data. It has close ties with the data science field as businesses use these insights the application provides to make important decisions. This dependency will only increase over time as machine learning algorithms get better with more use and more accurate.

Machine learning is widely used in many different types of applications. Common uses include:

  • Speech recognition
  • Customer service chat bots
  • Photo/Video analysis, recognition
  • Recommendation engines
  • Automated stock trading

Data can be split in three different types:

  • Numerical
  • Categorical
  • Ordinal

Numerical data, as the name suggests, is data consisting of numbers, which can be categorized as either discrete data or continuous data

  • discrete data - a numerical data type with countable elements, represented as real numbers. ex. number of students in a class
  • continuous data - a numerical data type with uncountable elements, represented as intervals on a real number line. They ca be either finite or infinite. ex. a person's height

"Categorical data is data that cannot be measured against each other and are classified by categories and aliases, eg. colors, yes/no values"[1]

Ordinal data is similar to categorical data, but it differs in that this data is measurable against itself. eg. school letter grades

Knowing the type of dataset you are working with will heavily impact your application's ability to "learn" new techniques for processing your data.

Source: [1] W3Schools - Machine Learning