ITS AI Glossary

What is an A/B Test?

• A method of creating a backup for testing environments.
• A randomized experiment comparing two variants to measure their effectiveness.
• A process for labeling datasets in machine learning.
• A testing approach exclusively for binary data.

What is Accuracy in AI?

• The precision of a machine learning model's predictions.
• The percentage of correct predictions made by a model.
• The measure of data imbalance in a dataset.
• The model's ability to generalize to unseen data.

What is Adversarial Testing in AI?

• Simulating user feedback to improve a model's accuracy.
• Testing a model by exposing it to deceptive or manipulative inputs.
• Analyzing decision trees for classification performance.
• Comparing models to find the best fit for a dataset.

What does AI stand for?

• Automated Information.
• Artificial Intelligence.
• Advanced Integration.
• Algorithmic Inference.

What is an Algorithm?

• A collection of data stored in a database.
• A set of rules or instructions used to solve a problem.
• A visualization tool for decision-making processes.
• A system for tracking computational errors.

What does 'Anomaly' mean in AI?

• A standard pattern or data point in a dataset.
• An unexpected or rare event that deviates from the norm.
• A feature of a supervised algorithm.
• A repetitive behavior observed in datasets.

What is an API?

• A data encryption method for secure communication.
• A standardized interface for software applications to communicate.
• A programming language used in machine learning.
• A database framework for AI systems.

What is an Assumption in AI modeling?

• A measure of how sensitive a model is to input changes.
• A hypothesis or premise used to build a model.
• A metric for evaluating data quality.
• A decision tree's root feature.

What is an Attack Surface?

• The area of memory a model uses during training.
• The collection of possible vulnerabilities in a system.
• The layers in a neural network.
• The dataset size required to train a model.

What is AUC (Area Under the Curve)?

• The length of time a model takes to converge.
• A metric used to measure the performance of a classification model.
• The difference between precision and recall.
• A curve that represents decision tree growth.