Computational Thinking and Artificial Intelligence Lecture Notes

A comprehensive set of vocabulary flashcards covering the core concepts of Computational Thinking, AI domains, and the AI project cycle as discussed in the lecture.

Debugging

The process of identifying and fixing errors within a program or system.

Generalization

The act of applying a solution to similar problems in different contexts.

Automation

Using machines to perform repetitive tasks to increase efficiency.

Computation

The process of systematically following a set of rules to transfer input data into a desired output.

Computational Thinking

A problem-solving approach that follows specific principles and techniques to handle complex problems, enabling logical and systematic thinking.

Decomposition

Breaking down complex tasks into smaller, more manageable parts or sub-problems to focus on solving each part separately.

Pattern Recognition

Identifying similarities, patterns, or trends within data to gain insights and meaningful information, which is essential for AI predictions and decision-making.

Abstraction

Creating simplified models or representations of a problem by removing unnecessary details while focusing only on essential aspects, such as a real-time map focusing only on roads and landmarks.

Algorithm Design

Developing a step-by-step plan or set of instructions to solve a problem, similar to a recipe.

Computer Vision (CV)

A domain of AI that uses visual data, including videos and images, to allow computers to understand visual information.

Natural Language Processing (NLP)

A domain of AI that focuses on textual data, enabling machines to understand, generate, and manipulate human language.

Data (AI Domain)

Techniques used to analyze, interpret, and draw insights from abstract numerical data.

Problem Scoping

The first stage of the AI project cycle which involves identifying the problem and stakeholders using the $4W$ framework: Who, What, Where, and Why.

Data Acquisition

The second stage of the AI project cycle involving the collection of authentic, reliable, and secure data that serves as the base for the AI project.

Data Exploration

The third stage of the AI project cycle which involves extracting useful information and insights from the gathered data.

Modelling

The fourth stage of the AI project cycle that involves choosing the specific models that suit the project requirements.

Evaluation

The fifth stage of the AI project cycle where models are tested in real-life situations.

Deployment

The final stage of the AI project cycle which involves deploying the solution to its intended environment.

Training Data

The portion of data, typically $70\%$, used to train an AI model through algorithms.

Testing Data

The portion of data, typically $30\%$, used to test the performance and accuracy of an AI model after training.

System-map

A tool used to find relationships between different elements of a system to help identify problems and reach project goals.