- Affiliate marketing
A type of relationship marketing where websites or organizations get a commission for either referring traffic or contributing sales for a particular brand.
Having human-like characteristics.
- Application programming interface (API)
Refers to the tools and communication protocols used to build software. An API is the building block by which computer programs and applications are built.
An arm is a limb on the upper part of a person's body. You can call the front limb of any animal an arm, though they're more often called "legs."
- Artificial general intelligence (AGI)
Also known as “superintelligence”, AGI refers to an intelligent system that encompasses a broad range of subjects, similar to or even surpassing average human intellect. AGI does not currently exist in the present.
- artificial intelligence
- Artificial Intelligence (AI)
Also called machine intelligence, this refers to the intelligence demonstrated by machines, as opposed to the natural intelligence inherent in humans and living organisms such as animals. AI can also be defined as a machine’s ability to interpret data and learn from it.
- Artificial intelligence marketing (AIM)
AIM is a new method of leveraging AI and customer databases to achieve marketing goals. By using machine learning, marketers can anticipate the customer’s next move, improve marketing and even predict future market trends.
- Artificial narrow intelligence (ANI)
ANI refers to intelligence that covers only a limited set of topics relevant to the machine’s function. Voice assistants and self-driving AI are examples of weak or narrow AI , where their intelligence is limited to their specific task.
- Augmented reality (AR)
Refers to an interactive experience, where the real-world environment is “augmented” by computer-generated information. This can come in the form of graphics, labels, data overlays or a graphical UI that augments physical reality. AR is distinguished from virtual reality (VR) in that AR augments the real-world environment, instead of supplanting it.
The process of replacing an activity previously performed by a human with a machine, either partially or fully.
- Autonomous computing
A characteristic whereby distributing computing resources are self-managed and can adapt to sudden changes.
“this award-winning bridge spans a distance of five miles”
- Bayesian networks
This is a type of statistical model that displays a set of variables and their conditional dependencies, using a directed acyclic graph. In the field of AI, Bayesian networks are used in algorithms to illustrate inference and machine learning.
- Behavior tree
A mathematical model used to describe the switchings between different tasks, in a modular way. Behavior trees are typically used in computer science, video game development, and robotics.
- Big data
This refers to the huge complex databases accumulated by companies. Such “big data” is often outside a human’s ability to analyze, and are typically processed using advanced software. This is where AI and machine learning come into play, in order to make sense of the data.
In Marketing, bots are often the external AI interface that interacts with the customer. Bots come in a variety of forms, from “chatbots” that respond to instant messaging inquiries, to “social bots” used to populate social media in support of a goal, brand or advocacy.
- Brute force
A kind of problem-solving technique that systematically tries all out possible combinations, hence using “brute force” to solve the problem. Commonly used in conjunction with hacking, where a brute force program tries all possible password combinations.
- Churn prediction modeling
Churn refers to the number of customers who end their contract. Churn prediction models are AI-built simulations that identify the factors and underlying causes behind churn, so that marketers can take proactive measures to reduce churn and keep customers engaged.
A derogatory term for deliberately sensational or misleading headlines (bait) for the purpose of attracting views (clicks).
A general term for the Internet. Can also refer to cloud-based storage, where data is stored on online servers, instead of locally through the device memory.
- Cluster analysis
A type of unsupervised machine learning that aims to find hidden patterns or grouping in a particular data set.
- Cognitive computing
A computerized model based on how the human brain thinks. Cognitive computing uses data mining, natural language processing as well as pattern recognition.
- Computer vision
This technology allows machines to recognize what they are seeing using facial and object recognition. Computer vision is increasingly being used in a number of industries, from self-driving cars and autonomous drones to social media apps like Snapchat.
- Concept drift
A condition where the statistical properties of the model being predicted change over time, so that predictions become less accurate as time passes.
- Content curation
Gathering information on a particular topic or area of interest. Pinterest is an example of curated content.
- Content generation
A new process where AI produces automated content in the form of text. An example is Wordsmith, a natural language generation engine that turns data into text.
- Content management system (CMS)
A system for the creation and modification of digital content, typically for the web. An example would be Wordpress, a popular CMS used for blogging and creating websites.
- Conversational UI
This refers to interaction by speech or text, as opposed to graphical UI.
In technology, cookies are small text files stored on web browsers. These are used by websites for authentication, to store user preferences, and any other data to help the browser access a specific server or website.
- Cross-channel marketing
A marketing tactic where customers are engaged across multiple channels, not necessarily digital. This can range from websites and TV to physical billboards, events, etc.
- Cross-device marketing
A cohesive marketing strategy or message that targets customers across multiple devices.
- Data analytics
Is the science of qualitative and quantitative processes used to examine data and draw meaningful insights. In today’s world of Big Data, AI and machine learning are becoming indispensable tools to make sense of the increasingly complex data that would overwhelm normal human marketers.
- Data efficiency
Refers to efficient practices for using data, such as when storing, accessing or filtering such volumes of data. Data efficiency is an important part of any enterprise, where efficient data storage and access can directly impact operating cost, revenue and manpower hours.
- Data mining
The process of discovering patterns and correlations within Big Data, often for practical business applications. For instance, a customer’s profile could be mined for interests, search history, or products bought.
- Deep learning
This type of machine learning is based on learning data representations, versus narrow algorithms. Deep learning is differentiated by its use of multiple layers of calculation and is often associated with neural networks. It can be supervised or unsupervised.
- digital advertising
- Digital assistants
Also known as voice assistants, these refer to the voice-controlled interfaces in an increasing number of gadgets. Examples include Amazon’s Alexa, Apple’s SIRI, Google Assistant, and Samsung’s Bixby, to name just a few.
- Dynamic pricing
A product strategy where price is determined based on demand, availability and the unique profile of the customer.
- Existential risk
A hypothesis where advances in AI could result in human extinction or global catastrophe. A popular example would be the Skynet scenario in The Terminator movie franchise.
- Expert systems
A computer system that embodies a stored body of knowledge on a specific topic, thus emulating a human expert in that specific subject. Expert systems are comprised of a knowledge base, and an inference engine that applies rules to the stored facts in order to deduce new facts.
In AI terms, a fluent is a condition that can change over time. Fluents can be represented as either functions or predicates.
- Fuzzy logic
A type of logic where variables can have partial truth, as opposed to just 1 (completely true) or 0 ( completely false).
A technique where people are persuaded to do something by promising a potential reward.
- Graph analysis
In AI, graphs are used to represent networks of interconnected datasets, such as people or organizations. Graph analysis is typically used by product recommendation engines to illustrate the relationships between customers and products.
- Graphical user interface (GUI)
GUIs are the visual interface for interacting with an object or website. Examples of GUIs are operating systems like Windows 10 for PCs, or iOS for Apple devices.
- Image recognition
This technology allows machines to understand what they are seeing, using a preprogrammed database to recognize people, objects, or places. Image recognition is crucial to computer vision and all the industries that rely upon it.
- Inference engine
A component of AI that stores logical rules. Inference engines are used in conjunction with the knowledge base to deduce new facts.
- Intelligent agent
A machine that is able to perceive changes in its environment and act accordingly. An example could be sensor-equipped gadgets that react to touch, temperature or lighting conditions.
- Internet of Things (IoT)
IoT refers to everyday objects that are connected to the Internet, such as smart homes, appliances and even smart plugs that provide Internet connectivity to otherwise “dumb” household objects.
- Invisible brand
A term coined in the book “The Invisible Brand: Marketing in the Age of Automation, Big Data, and Machine Learning”, this refers to an emerging class of market forces that influence individual consumers through the hidden hand of personalized information and persuasion, coupled with new technologies like machine learning and natural language processing.
- Knowledge engineering
A subfield of AI, this discipline tries to recreate the behavior and thought process of a human who is an expert in a specific field. Knowledge engineering aims to create expert systems, using an expandable knowledge base coupled with a rules engine so that stored knowledge can be applied to specific situations.
- Machine learning
Refers to the scientific study of algorithms and statistical models uses by machines to progressively improve their performance on a given task. When applied in a business environment, machine learning can be considered predictive analytics.
- Mobile first
A website design practice where mobile users are prioritized first over other devices.
- Natural language generation
This technology transforms data into written text or vocal speech. This can be something as simple as pish notifications and artificially generated newsbriefs to personalized Alexa flash bulletins.
- Natural language processing
A machine’s ability to recognize human speech. Natural language platforms include Amazon’s Lex, IBM’s Watson and Google’s api.ai.
- Near field communication (NFC)
A communications protocol that allows electronic devices to communicate with one another across a very short range. Examples include the NFC technology in smartphones that allow contactless payments in NFC-enabled retail outlets.
- Neural network
This refer to computing systems that mimic the structure of biological brains. A neural network is composed of many different algorithms working together in the style of neurons and nodes to process complex data.
- Predictive analytics
A branch of analytics concerned with making predictions about future events, through data mining, modeling, statistics, and machine learning.
- promoted pins
Shorthand for “psychological technology”, this refers to new marketing innovations that persuade people at a personalized level, using voice-based conversations with machines that are able to learn.
- Sentiment analysis
A new type of “opinion-based” data mining where text is processed based on context, in order to identify the user’s attitude on a particular topic, product or situation.
- Supervised learning
This type of machine learning requires constant human supervision throughout the learning process using labeled data sets to build knowledge.
- Technological singularity
A hypothesized event where the creation of an artificial superintelligence triggers runaway technological growth, and results in dramatic changes to human civilization.
- Turing Test
The test of machine’s ability to display behavior indistinguishable from that of a human. The machine is considered to have passed if the human participant is unable to distinguish if the machine he/she is interacting with is human or artificial.
- Uniform resource locator (URL)
The web address for a specific website or webpage, commonly typed in the address bar of web browsers.
- Unstructured data
Collected data that is not organized in a specific way, such as social media posts, as opposed to a neatly ordered Excel sheet.
- Unsupervised learning
This type of machine learning is distinguished from supervised learning by not requiring constant human supervision. In unsupervised learning, the data isn’t labeled and the machine is left to identify patterns or correlations to make sense of the data and come up with an answer.