AI unit

method that employs a practical solution (rather than a theoretical one) to a problem; when applied to algorithms this includes running tests and obtaining results by trial and error.

Answer explanation

The method described is a heuristic, which focuses on practical solutions and involves trial and error to solve problems, particularly in algorithms. This distinguishes it from purely theoretical approaches.

statistical measure used to make predictions from data by finding learning relationships between the inputs and outputs.

• Random guessing

• Pattern recognition

• Trial and error

• Regression analysis

Answer explanation

Regression analysis is a statistical measure that helps in making predictions by identifying relationships between input variables and output outcomes, making it the correct choice for the question.

system which is able to identify hidden patterns from input data – the system is not trained on the ‘right’ answer.

• Supervised learning

• Reinforcement learning

• Semi-supervised learning

• Unsupervised learning

Answer explanation

Unsupervised learning identifies hidden patterns in data without labeled responses. Unlike supervised learning, it does not rely on 'right' answers, making it suitable for discovering structures in unlabeled datasets.

data where we know the target answer and the data object is fully recognised

• Labelled data

• Reward and punishment

• Unlabelled data

• Back propagation

Answer explanation

Labelled data refers to datasets where each data object is associated with a known target answer, making it essential for supervised learning tasks. This distinguishes it from unlabelled data, which lacks such annotations.

​ (a)   – system which is able to predict future outcomes based on past data. It requires both input and output values to be used in the training process.

​ (b)   – system which is able to identify hidden patterns from input data – the system is not trained on the ‘right’ answer.

​ (c)   – system which is given no training – learns on basis of ‘reward and punishment’.

​ (d)   – system that interactively queries source data to reach the desired result. It uses both labelled and unlabelled data, but mainly unlabelled data on cost grounds.

• Semi-supervised (active) learning

• Reinforcement learning

• Unsupervised learning

• Supervised learning

• Machine learning

• Reward and punishment

• Unlabelled data

Answer explanation

Supervised learning predicts outcomes using past data with input and output values. Unsupervised learning finds hidden patterns without labeled answers. Reinforcement learning learns through rewards and punishments. Semi-supervised learning uses both labeled and unlabeled data.

Match the following

• A* algorithm
• an algorithm that finds the shortest route between nodes or vertices but uses an additional heuristic approach to achieve better performance than Dijkstra’s algorithm.
• Dijkstra’s algorithm

• fundamental unit from which graphs are formed (nodes and vertices are the points where edges converge)

• Node or vertex

• an algorithm that finds the shortest path between two nodes or vertices in a graph/network.

• Machine learning

• systems that learn without being programmed to learn

internet bot that systematically browses the world wide web to update its web page content.

• Web crawler

• Search engine

• Web scraper

• Data miner

Answer explanation

A web crawler is an internet bot that systematically browses the web to update content on web pages. It differs from search engines, web scrapers, and data miners, which have different functions.

networks of interconnected nodes based on the interconnections between neurons in the human brain. The system is able to think like a human using these neural networks, and its performance improves with more data

• Vascular networks

• Circuit networks

• Neural networks

• Muscle networks

Answer explanation

The question describes a system that mimics human thought processes through interconnected nodes, which aligns with the definition of neural networks. These networks improve their performance with more data, confirming that 'Neural networks' is the correct answer.

computer program set up to simulate conversational interaction between humans and a website

• Spreadsheets

• Regression

• Web crawler

• Chatbot

Answer explanation

A chatbot is a computer program designed to simulate conversational interaction with users, making it the correct choice for the given description. Other options like spreadsheets and web crawlers do not serve this purpose.

method used in artificial neural networks to calculate error gradients so that actual node/neuron weightings can be adjusted to improve the performance of the model.

• Upwardpropagation

• Backpropagation

• Sidewayspropagation

• Forwardpropagation

Answer explanation

The correct method is Backpropagation, which calculates error gradients in neural networks. This process allows for the adjustment of weights in neurons to enhance model performance, making it essential for training neural networks.