Class IX Artificial Intelligence - Unit 1: AI Reflection, Project Cycle and Ethics

Vocabulary flashcards covering core definitions, types of AI, domain areas, practical applications, project cycle stages, modeling approaches, evaluation metrics, and ethics based on Class IX AI lecture notes.

John McCarthy

The Dartmouth Assistant Professor who introduced the term Artificial Intelligence in 1956.

Artificial Intelligence (AI)

A general term referring to hardware or software that exhibits behavior that appears intelligent, or the simulation of human intelligence in machines programmed to think and mimic human actions.

AI Formula

The essential combination required to create an intelligent machine: Data + Algorithm = AI Machine!

Machine Learning (ML)

A subset of artificial intelligence that enables computer programs to automatically learn, adapt, and improve at tasks with experience without human assistance.

Deep Learning (DL)

A subfield of machine learning that enables software to train itself on vast amounts of unstructured data (such as text, images, or videos) to develop algorithms independently.

Weak AI (Narrow AI)

A type of AI trained to perform a dedicated task with intelligence within pre-defined limitations, such as Apple's Siri, IBM's Watson, or chess-playing software.

General AI

A type of AI that could perform any intellectual task with efficiency like a human and think independently; no such system currently exists.

Super AI

A hypothetical level of system intelligence that surpasses human intelligence across any cognitive task, including reasoning, planning, and judgment.

Reactive Machines

The most basic type of functional AI that does not store memories or past experiences, focusing solely on current scenarios (e.g., Google's AlphaGo).

Limited Memory AI

AI systems that store past experiences or temporary data for a short period of time to navigate current tasks, such as self-driving cars.

Theory of Mind AI

A non-existent, developing type of AI designed to understand human emotions, beliefs, and social interactions.

Self-Awareness AI

A hypothetical future level of AI that possesses its own consciousness, sentiments, and self-awareness, exceeding human cognitive capacity.

Computer Vision (CV)

An AI domain that processes images and videos, enabling machines to interpret and understand visual information.

Natural Language Processing (NLP)

An AI domain focused on textual data that enables machines to comprehend, generate, and manipulate human language.

Statistical Data Domain

An AI domain that utilizes statistical techniques to analyze, interpret, and extract insights from numerical or tabular data.

Face Lock in Smartphones

A Computer Vision application where the front camera captures facial features during initiation and matches them to unlock the device.

Smart Assistants

Applications like Apple's Siri and Amazon's Alexa that recognize patterns in human speech, infer meaning, and deliver appropriate responses.

Fraud and Risk Detection

A financial application of AI that analyzes customer profiling, past expenditures, and essential variables to evaluate default probability and manage risk.

Medical Imaging AI

A computer-supported application that converts 2D medical scan images into interactive 3D models to assist doctors in health diagnosis.

AI Project Cycle Stages

The sequential stages involved in an AI project: Problem Scoping, Data Acquisition, Data Exploration, Modeling, Evaluation, and Deployment.

Problem Scoping

The first stage of the AI project cycle involving problem identification, goal setting, parameter observation, and applying the 4Ws problem framework.

4Ws Problem Canvas

A problem-scoping framework addressing Who is affected, What is the nature of the problem, Where it arises, and Why it is worth solving.

Data Acquisition

The second stage of the AI project cycle focused on collecting relevant facts, statistics, and information required for the project.

Training Data

The dataset collected and fed into an AI system to train the machine and teach it how to learn or make predictions.

Test Data

The processed dataset separated from the acquired data used to test and evaluate the accuracy and efficiency of a trained model.

Data Exploration

The third stage of the AI project cycle where data is analyzed and visualized using bar graphs, pie charts, or histograms to discover trends and patterns.

Modeling

The fourth stage of the AI project cycle involving the process of translating data trends and patterns into mathematical representations or AI algorithms.

Rule Based Approach

An AI modeling approach in which explicit rules and instructions are pre-programmed into the machine by the developer.

Learning Based Approach

An AI modeling approach where the machine dynamically learns from data changes by generating and adapting its own algorithm.

Evaluation Stage

The fifth stage of the AI project cycle that tests model reliability and calculates performance using metrics such as Accuracy, Precision, Recall, and F1 Score.

Deployment Stage

The final stage of the AI project cycle where validated AI models are integrated, monitored, and implemented in real-world scenarios.

Morals

Societal beliefs and standards regarding right and wrong behavior that vary between different cultures and societies.

Ethics

Individual guiding principles and chosen values used to evaluate what is good or bad in specific situations.