New & Next

12 Artificial Intelligence Terms that Every Marketer Should Know


Artificial intelligence is transforming the marketing industry. Whether you are an experienced marketer or new to the industry, there are a number of ways you can use it to enhance the results of your marketing campaigns.

Before you even start researching the available AI technologies, though, you need to start with the basics: the vocabulary.

Here are 13 terms every marketer should know.

1. Object recognition

Object recognition, which includes facial recognition, refers to an application or a program’s ability to detect certain objects, faces, and so on.

More advanced applications can even detect and mimic behavior and raise a red flag when something or someone is out of line.

In recent years, many countries have implemented facial recognition technologies into their customs departments to easily identify people, read their gestures, and detect criminals.

2. Data mining

Artificial intelligence implementations need data—lots of it. Data mining refers to the process of filtering that large amount of information the application needs to be able to function and give precise estimations.

Let’s say a facial recognition app needs 100,000 examples of people’s faces to be able to “learn” how to properly identify people. The process of transforming that raw information it receives into information it can make use of is data mining.

One of the most infamous examples of data mining was the Trump campaign’s use of it during the 2016 U.S. presidential election. Together with Cambridge Analytica, his campaign team mined large amounts of data about his voters using Facebook. They used that data to pinpoint citizens who would consider voting for him, targeting them with paid advertising on social networks.

3. Neural network

Neural networks refer to AI’s ability to solve complex problems, just like humans do. They split a problem into smaller pieces, which are then solved to achieve a result.

4. Chatbot

A chatbot is an artificial intelligence-based application that simulates a human-to-human conversation. It studies a user’s input, identifies keywords and replies according to that given input.

If implemented in an e-commerce website, for example, it could give the user information about certain products, different product categories, it could give useful links to customers, and so on.

Chatbots can lead to great results, including an increase in sales.

5. Deep learning

Deep learning is another layer on top of the neural network, similar to “learning from experience” in human behavior.

As we said, an AI application needs lots and lots of information to function properly. From each example it receives, it learns more about the topic it studies.

6. Natural language processing

Natural language processing is the process of understanding and processing a human’s input.

For example, a chatbot would need to understand and process a question that the user asks through the chat window. For artificial intelligence applications to understand this input, they need to understand how humans speak.

7. Turing tests

The Turing test was invented by computer science pioneer Alan Turing in the 1950s. It refers to a machine’s ability to perform intelligent behavior that is similar or identical to a human being.

An example of a Turing test would be to send to a chatbot the following text: “I want more information about iPhone 10.” If the bot can process your input and send you to the iPhone 10 information on Apple’s website, that would be a successful result.

8. Machine learning

Machine learning is all about combining big data, data mining, neural networks, natural language, and many more techniques in order to create a machine or application that can learn from the input it gets from either a large data set or the user it interacts with. Basically, it’s a subdomain of artificial intelligence and an umbrella for various processing techniques.

9. Black box storing

Black box information storing is the process of storing information about users, for example, without actually knowing their personal, sensitive information. For example, Google, through cookies, stores information about your online behavior, such as which websites you visit, when, and how often. It  recognizes you by your cookie id, not your personal information.

10. Supervised learning

Supervised learning is the process through which a machine is given both the input that it should process and the result it should return.

It’s like telling your son: “Here’s the supermarket, here’s the shopping list, I’ll wait for you outside.” He knows where to go and what he should come back with, but he doesn’t know where the products are inside the store.

11. Algorithm

Algorithms are the foundation of artificial intelligence. They are responsible for processing the information an AI application “ingests.” These applications are built using thousands of thousands of different pieces, each one being able to do one thing. They are then combined in order to come up with the desired result.

12. Behavior analysis

The most basic example of behavior analysis is when Google studies your online activity. It takes into consideration multiple variables, such as age, websites you visit, places you visit, places you like, and so on, in order to analyze your behavior and create a pattern. Then, it is able to target you with ads according to that pattern.


Artificial intelligence evolves more each day. In the marketing industry, it could be the make-or-break technology for your business. Before you start researching how to use it, make sure you understand the terms.

Alice Jones

Alice Jones is a technical writer based in San Francisco.

Post a Comment

Your email address will not be published. Required fields are marked *