Unit 1 – AI Fundamentals & Project Cycle Vocabulary

Vocabulary flashcards summarising the key terms, concepts, applications, and stages presented in Unit 1 on AI fundamentals and the AI Project Cycle.

Artificial Intelligence (AI)

The capability of a machine to mimic human traits—collecting data, learning, making decisions, predicting outcomes, and self-improving.

AI Domain

A specific area within AI defined by the nature of data it handles, such as Natural Language Processing, Computer Vision, or Statistical Data.

Data for AI

Historical or real-time records (numbers, text, images, etc.) that an AI system uses to learn patterns and make predictions.

Natural Language Processing (NLP)

AI domain that enables machines to comprehend, generate, and manipulate human language in textual or spoken form.

Computer Vision (CV)

AI domain focused on enabling machines to interpret and understand images and video like humans do.

Statistical Data (AI Context)

Numeric or tabular information examined with statistical techniques to extract insights for AI models.

Smart Assistant

Voice-controlled software (e.g., Siri, Alexa) that recognizes speech patterns, infers meaning, and responds helpfully.

Face Lock

Smartphone security feature that uses computer-vision algorithms to match a user’s facial features for authentication.

Fraud & Risk Detection

AI application in finance that profiles customers and analyses past expenditures to predict default or fraudulent activity.

Medical Imaging

AI-aided analysis that converts 2-D scans into interactive 3-D models, assisting doctors in diagnosis and treatment planning.

Rock, Paper & Scissors AI

Online game demonstrating ‘Data for AI’ where an algorithm learns a player’s move patterns to predict and win.

Semantris

Google word-association game showcasing NLP; the AI ranks words by semantic similarity to the player’s clues.

Quick, Draw!

Google doodling game using computer vision; a neural network guesses in real time what the user is drawing.

AI Project Cycle

Cyclical framework outlining the stages for building an AI solution: Problem Scoping, Data Acquisition, Data Exploration, Modelling, Evaluation, and Deployment.

Problem Scoping

First stage of the AI Project Cycle where the issue is identified, stakeholders studied, and a clear goal is set.

Data Acquisition

Stage of gathering relevant, sufficient, and diverse data needed for training and validating an AI model.

Data Exploration

Process of cleaning, visualizing, and discovering patterns or trends in acquired data to guide modelling choices.

Modelling

Stage where appropriate algorithms are trained on processed data to learn patterns and make predictions.

Evaluation

Assessing a trained model’s accuracy and reliability, often by comparing its predictions against a test dataset.

Deployment

Releasing the validated AI model into real-world use, integrating it within apps, devices, or services.

4Ws Problem Canvas

Problem-scoping tool that captures Who is affected, What the problem is, Where it occurs, and Why solving it matters.

Stakeholders

Individuals or groups directly or indirectly affected by the problem and likely to benefit from its solution.

Problem Statement Template

Structured sentence summarizing stakeholders, issue, context, and desired benefits to keep projects goal-focused.

Goal Statement

Concise description of what an AI project aims to achieve, often starting with ‘How might we…’

Theme (AI Project)

Broad area of interest (e.g., Health, Agriculture) from which more specific project topics are chosen.

Topic (AI Project)

Focused subject within a theme (e.g., pest control under Agriculture) used to narrow project scope.

CottonAce App

Mobile AI application that helps Indian cotton farmers detect pest infestations, optimize pesticide use, and boost profits.

Pink Bollworm

Pest that commonly infests cotton crops; target organism for the CottonAce AI detection system.

AI Project Cycle Mapping

Exercise of linking each practical project activity to its corresponding stage in the AI Project Cycle.

AI vs. IT Project

AI projects revolve around learning from data and iterative model improvement, whereas traditional IT projects follow fixed rules and logic.

Sustainable Development Goals (SDGs)

UN-defined global objectives that can inspire AI project themes addressing societal and environmental issues.

Non-AI Solution

Traditional method of solving a problem without machine learning, considered during initial brainstorming.

Evidence Collection

Gathering articles, statistics, or testimonials to prove that a chosen problem genuinely exists and is worth solving.

Context (Problem Scoping)

Specific situation or environment in which the identified problem occurs, critical for designing relevant solutions.