AI Problem Definition
Intelligent programmed to mimic human thinking
What is AI?
Data represented as 1s and 0s.
What is Binary?
A correct positive prediction
What is a True Positive?
Allows applications to communicate.
What is an API?
Comparing two versions of a system.
What is an A/B Test?
A set of rules used to solve problems.
What is an Algorithm?
The kind of data (text, number, date)
What is Data Type?
A missed positive result.
A: What is a False Negative?
A file format using key-value pairs.
What is JSON?
Shutting down an AI system.
What is Decommission?
When data unfairly favors a group.
What is Bias?
An input used in an AI model.
What is a Feature
When a model memorizes training data but fails on new data
What is Overfitting
Steps that transform data into predictions.
What is a Pipeline?
A record of system activity
What is a Log?
A model that sorts into categories like yes/no.
What is Classification?
A list of numeric values describing data
What is a Feature Vector
A chart showing prediction outcomes
What is a Confusion Matrix
When the meaning of data changes over time.
What is Concept Drift?
Measures how much users interact with a system.
What are Usage Metrics?
This type of machine learning uses labeled data and predicts outcomes using binary dependent variables.
What is Logistic Regression?
This process represents categorical variables as binary vectors so AI models can interpret them numerically.
What is One-Hot Encoding?
This metric measures the proportion of variance in a dependent variable that can be predicted from independent variables.
What is R-Squared?
This mechanism identifies shifts in underlying data that may reduce the accuracy of an AI solution over time.
What is a Drift Detector?
This mode occurs when an AI model continues functioning but no longer performs at maximum capability due to performance issues or drift.
What is Degraded Mode of Operation?