AI Glossary
AI Glossary: Top Terms Explained
Welcome to the AI Glossary section! Here you will learn key terms and concepts related to AI. Understanding these terms will give you the confidence in this AI world. Let’s get started!
📘 Artificial Intelligence (AI)
Definition
AI refers to machines or software that can perform tasks that usually require human intelligence, such as understanding language, recognizing images, or making decisions.
Real-World Example
Virtual assistants like Siri or Alexa interpret your voice commands to play music or set reminders—showcasing AI by understanding speech and responding appropriately.
📘 Machine Learning (ML)
Definition
A subset of AI where computers learn from data and experiences rather than being explicitly programmed for each task.
Real-World Example
Recommendation systems on Netflix or YouTube learn what you like to watch and suggest videos or shows based on your previous viewing habits.
📘 Deep Learning
Definition
A specialized branch of machine learning that uses layers of algorithms called “neural networks” to mimic how the human brain processes information.
Real-World Example
Facial recognition on social media platforms automatically tagging your friends in photos by learning patterns in their faces over time.
📘 Neural Networks
Definition
Computational models inspired by the human brain. They have “neurons” arranged in layers that process information and learn patterns.
Real-World Example
A spam-filtering system in your email that “learns” to distinguish between spam messages and legitimate emails.
📘 Natural Language Processing (NLP)
Definition
A field of AI focused on understanding and generating human language.
Real-World Example
Auto-correct and predictive text on your smartphone interpret what you type and suggest words or correct spelling in real-time.
📘 Computer Vision
Definition
AI that enables computers to interpret and understand visual information from the world, such as images or videos.
Real-World Example
Self-driving cars detect traffic lights, pedestrians, and road signs to navigate safely.
📘 Supervised Learning
Definition
A type of machine learning where the model is trained on labeled data (i.e., the correct answers are provided).
Real-World Example
A system that can identify whether an email is “Spam” or “Not Spam” after being trained on thousands of examples labeled as spam or not spam.
📘 Unsupervised Learning
Definition
A type of machine learning where the model tries to find patterns in data without any labels or predefined “right answers.”
Real-World Example
Grouping customers by purchasing habits without telling the system which group they belong to in advance. Companies then target each group with specialized offers.
📘 Reinforcement Learning
Definition
A learning technique where an AI “agent” learns by trial-and-error, receiving rewards or penalties for actions, and optimizing its behavior over time.
Real-World Example
AI-powered game bots that learn how to win at chess or video games by constantly playing and adjusting their moves based on wins/losses.
📘 Algorithm
Definition
A set of instructions or rules a computer follows to solve a problem or perform a calculation.
Real-World Example
A recipe for baking a cake—if you follow the steps in the correct order, you get the desired result.
📘 Model (AI Model)
Definition
The final “trained” product of a machine learning process that can make predictions or decisions.
Real-World Example
After teaching a computer to recognize cats vs. dogs with thousands of labeled pictures, the resulting program is the “model” that can classify new images.
📘 Training Data
Definition
The information (text, images, numbers) used to teach an AI model to make predictions or decisions.
Real-World Example
To teach an AI to recognize handwritten digits, you provide it with thousands of samples of handwritten numbers along with their correct labels.
📘 Bias (in AI)
Definition
When an AI system produces results that are prejudiced due to flawed data or assumptions.
Real-World Example
If an image recognition model is primarily trained on photos of light-skinned faces, it may perform poorly on darker-skinned faces.
📘 Overfitting
Definition
When an AI model learns the training data too well—like memorizing the answers—causing it to perform poorly on new, unseen data.
Real-World Example
A student who memorizes a practice test’s exact questions and answers might do well on that test but fail a different exam covering the same subject.
📘 Underfitting
Definition
When an AI model doesn’t learn enough patterns from the training data, resulting in poor performance.
Real-World Example
A student who barely studies and struggles with both practice tests and new exams.
📘 Inference
Definition
The process of using a trained AI model to make a prediction or decision on new data.
Real-World Example
When a photo app identifies your face in a new selfie, it’s “inferring” based on what it learned during training.
📘 Classification
Definition
A task where an AI model assigns labels to data (e.g., “Dog” or “Cat,” “Spam” or “Not Spam”).
Real-World Example
Sorting incoming emails into categories like “Work,” “Promotions,” or “Social.”
📘 Regression
Definition
A task where an AI model predicts continuous values (e.g., prices, temperatures).
Real-World Example
Predicting stock prices for next week or estimating the value of a car based on its mileage.
📘 Clustering
Definition
Grouping data that share common patterns or characteristics without prior labels.
Real-World Example
Grouping your photo library by faces or events, even if you haven’t labeled who’s in the photo or when it was taken.
📘 Data Mining
Definition
The process of finding patterns, trends, or relationships in large sets of data.
Real-World Example
Online retailers analyzing purchase histories to discover that many people buy milk and bread together, prompting store layouts or combos.
📘 Big Data
Definition
Extremely large and complex data sets that traditional software can’t easily handle.
Real-World Example
Social media companies collect billions of posts, likes, and comments daily to analyze trends or recommend content.
📘 Chatbot
Definition
A computer program designed to simulate conversation with human users.
Real-World Example
Customer service chat on a website that responds to your queries about order status or product returns.
📘 API (Application Programming Interface)
Definition
A set of rules and protocols for building software applications and allowing different programs to interact.
Real-World Example
A weather app on your phone uses a public weather service’s API to request today’s forecast and show it to you in a user-friendly format.
📘 Edge AI
Definition
Running AI algorithms directly on devices (phones, cameras, sensors) rather than sending data to the cloud for processing.
Real-World Example
Your smartphone can unlock by recognizing your face without needing to send images to a server—faster and more private.
📘 Cloud Computing
Definition
Using remote servers on the internet to store, manage, and process data instead of doing everything on your local computer.
Real-World Example
Google Docs stores your documents on Google’s servers, and you can access them from any device with an internet connection.
📘 Internet of Things (IoT)
Definition
A network of physical devices—like home appliances, cars, wearable tech—that are connected to the internet, collecting and sharing data.
Real-World Example
A smart thermostat learns your heating preferences and adjusts temperatures automatically, saving energy and money.
📘 Ethics in AI
Definition
Concerned with making AI systems fair, transparent, and safe, ensuring they benefit society without causing harm.
Real-World Example
Ensuring facial recognition isn’t misused for illegal surveillance or that automated hiring systems don’t discriminate against certain groups.
📘 Explainable AI (XAI)
Definition
AI techniques designed to make the model’s decisions understandable to humans.
Real-World Example
If an AI denies you a loan, XAI provides a clear explanation—such as a low credit score or insufficient income—rather than a “black box” answer.
📘 Generative AI
Definition
AI that creates new content—like text, images, or music—based on patterns it has learned.
Real-World Example
Tools like ChatGPT (text generation) or DALL·E (image generation) can produce original text passages or artwork based on user prompts.
📈 Challenges
- What is Natural Language Processing (NLP)?