Our AI in education glossary demystifies terminology related to the rapidly growing field of artificial intelligence.
Welcome to D2L’s AI in Education Glossary, where we aim to unravel the intricacies of artificial intelligence terminology. Whether you’re a teacher, curriculum designer or student, our glossary covers everything from neural networks to big data. Wherever possible, we’ve tried to share examples of how these terms are used in educational settings.
AI (Artificial Intelligence): the ability of a digital computer or computer-controlled robot to perform tasks commonly associated with intelligent beings, even though they may not be able to fully replicate human flexibility and understanding. These tasks may include reasoning, discovering meaning, generalizing, and learning from past experiences.
In simpler terms, AI is intelligence exhibited by machines, particularly computer systems.
Some examples of how artificial intelligence is used in education include:
Algorithm: a process or set of rules to be followed in calculations or other problem-solving operations, especially by a computer.
In AI, algorithms consist of instructions guiding computers in learning and autonomous operation. They form a subset of machine learning, enabling computers to perform calculations and tasks that would typically demand human intelligence
API (Application Programming Interface): an API is a set of rules that allows software applications to communicate and exchange data, features, and functionality. It simplifies development by enabling integration with other applications and acts as a bridge between different software components
Big Data: in the context of AI, big data refers to vast and complex datasets used to train and power AI systems. These datasets enable AI models to make data-driven decisions and predictions.
An example of big data and AI in education can be seen in Massive Open Online Courses (MOOCs). MOOCs generate extensive data on student interactions, completion rates and learning patterns. AI can then analyze this data to identify areas where students are struggling, make improvements to course design and enhance student engagement and retention.
Chatbot: an automated program or AI system designed to engage in conversations with users, often through text or voice, providing information, answering questions, or assisting with tasks.
Chatbots can be used in education on university or school websites to answer commonly asked questions pertaining to student programs, experiences, administration, fees and so much more.
ChatGPT: ChatGPT is an AI-powered language model developed by OpenAI, capable of generating human-like text based on context and past conversations.
Data Science: the process of analyzing large datasets to discover patterns and gain insights that inform decision-making.
In education, data scientists will often analyze large sets of student data to draw patterns and then recommend ways to increase retention, enrollment and more.
Deep Learning: a specialized subset of machine learning that uses neural networks with three or more layers. These neural networks attempt to simulate the behavior of the human brain—although they fall far short of matching its ability—to “learn” from large amounts of data.
Emergent Behavior: refers to actions or patterns that weren’t explicitly programmed into an AI system but developed as a natural outcome of its complexity and interactions.
Generative AI (GenAI): refers to a category of machine learning models that utilize training data to produce novel content. This content can include images, text, videos, audio and other forms of media. GenAI models leverage neural networks to recognize patterns within existing data and generate new content based on what they are trained on.
Some examples of generative AI in education include:
Hallucination: AI hallucination refers to situations where an AI model produces outputs that diverge from reality. Recognizing the reasons behind this phenomenon and understanding its impact are essential for responsible AI implementation.
Hyperparameter: a hyperparameter is a configuration set externally before training in machine learning. Unlike model parameters, which are learned during training, hyperparameters guide the learning process but are not adapted from the data.
Large Language Model (LLM): an advanced deep-learning algorithm that uses many parameters and training data to understand and predict text. These generative artificial intelligence models can perform various natural language processing tasks beyond simple text generation, including revising and translating content.
Machine Learning: is a field within AI and computer science that aims to enable AI systems to learn from data and algorithms, mimicking human learning processes. It achieves this by analyzing extensive datasets without requiring explicit programming instructions.
Some examples of machine learning in education include:
Natural Language Processing (NLP): NLP combines computational linguistics—which involves rule-based modeling of human language—with statistical and machine learning models to enable computers and digital devices to recognize, understand, and generate text and speech.
Neural Network: a neural network is a model that mimics the way biological neurons work together to make decisions. It consists of interconnected units called neurons that send signals to one another, allowing it to perform complex tasks. In essence, neural networks process information like the human brain, using layers of artificial neurons to identify patterns and reach conclusions.
OpenAI: a research organization aiming to create safe and beneficial artificial general intelligence (AGI). They’ve developed powerful language models like GPT-4 and continue to advance AI research.
Pattern Recognition: pattern recognition in AI refers to machines’ ability to identify patterns in data, enabling them to make predictions and categorize information. It plays a crucial role in modern artificial intelligence systems, allowing them to learn from data and generalize patterns for practical use.
The language leaning app Duolingo is an example of how AI pattern recognition is being used in education. Duolingo uses data, pattern recognition and machine learning to improve its language courses. By analyzing user interactions and the nuances of various languages, Duolingo can optimize its exercises and adapt content accordingly.
Predictive Analytics: predictive analytics in AI involves using historical data and statistical algorithms to make predictions about future events or outcomes. In education, it helps schools and post-secondary institutes anticipate trends, identify patterns and optimize decision-making processes.
Prescriptive Analytics: prescriptive analytics in AI goes beyond predicting future outcomes. It recommends specific actions or decisions to optimize results based on historical data and models, helping organizations make informed choices.
Prompt: in the context of AI, a prompt refers to an input or query provided to a language model or other AI system. It serves as the starting point for generating a response, whether it’s in the form of text, code, or other output.
AI prompts have exploded in education and will play a significant role in the future. Good prompts help students engage with AI systems and enhance their learning experiences. They can also help teachers save time.
One example of how an AI prompt could be used in education is for designing a lesson plan. An educator could input the type of lesson and include parameters such as length, structure, types of questions etc., into a generative AI tool and obtain a plan they could then refine as they see fit.
Quantum Computing: enables AI systems to process vast amounts of data and execute complex algorithms at unprecedented speeds, leading to transformative advancements in AI applications.
Reinforcement Learning: an area of machine learning where an agent interacts with an environment, learning to take actions that maximize cumulative rewards. It involves balancing exploration and exploitation to optimize decision-making processes in dynamic scenarios.
Sentiment Analysis: involves analyzing text or data to determine the emotional tone or sentiment expressed. It helps classify whether a piece of content (such as a review, tweet, or customer feedback) is positive, negative, or neutral.
Sentiment analysis can play a pivotal role in education. Here are two ways it can be used:
Structured Data: in the context of artificial intelligence, structured data refers to data that follows predefined data models, has a consistent and clear structure, and can be easily accessed and processed by both humans and computer programs.
Supervised Learning: in AI, supervised learning refers to data that adheres to predefined data models, has a consistent and clear structure, and can be easily accessed and processed by both humans and computer programs.
Token: an AI token represents the smallest unit into which text data can be broken down. These tokens allow language models to efficiently process and interpret text, aiding in tasks like natural language understanding and generation.
Training Data: serves as the initial dataset used to train machine learning algorithms. It provides the foundation for creating and refining rules within these models. Whether labeled (with meaningful tags) or unlabeled (raw data), training data enables AI models to learn patterns and relationships, ultimately allowing them to make accurate predictions or perform specific tasks.
Transfer Test: a transfer test in AI refers to evaluating how well a pre-trained model, developed for one task, performs when applied to a related but distinct task. It assesses the effectiveness of knowledge transfer from the source task to the target task.
Join us as we explore the practical applications of AI for educators and instructional designers, with real-world examples of how you can use tools like ChatGPT, Google Bard and DALL-E to design better online learning experiences.
Written by:
Educators and training pros get our insights, tips, and best practices delivered monthly
