Digital Information Technology Final Exam Review Part 3

For Period 1, Digital Information Technology Final Exam 5/20/2025 For A or 90% and Higher.

Why is it important for iterative adjustments to be well-documented?

It's important for iterative adjustments to be well-documented as to see which adjustments made the model more accurate and which adjustments reduced the accuracy or had no impact at all for the machine learning or AI model.

What are explainability requirements?

Explainability requirements are requirments to be met for a project of a machine learning or AI model to have its results and outcomes be explainable or interpretable.

For which type of app is speed a more significant factor?

The type of app which speed is a more significant factors are any applications who use live data or real time data. Whether apps like digital maps/gps, social media, or even online commerce apps.

Using visualizations to show training model data can help identify:

Any outliers,missing data,corrupted data, or any other anomalies of that sort.

What is often the reason for seeing one or more outliers in data that do not follow a trend?

Often the reason for seeing one or more outliers in data that do not follow a trend is of the data set itself which has not been prepared properly leading to these outliers or anomalies.

What are visualizations?

Visualizations are tools to summarize and simplify outcomes or the model/data set itself. These can be charts,graphs, or models.

How might cost factor into developing an AI model?

Cost might factor into developing an AI model as this might change where the machine learning or AI model will be constructed in whether in physical servers or cloud based servers.

What are three common metrics used to evaluate AI models?

Three common metrics used to evaluate an AI model are precision,recall, and F1 score.

What is a test or validation data set?

A test or validation data set deals with evaluating or assesing the AI model to see if it's accurate with other data sets other than the training data set.

What is overfitting?

Overfitting occurs when a machine learning or AI model does great on the training data set but does not as greate with any other data sets.

What is underfitting?

Underfitting occurs when a machine learning or AI model does not properly intake information from the training data or any other data set causing it to lead to falty or inaccurate outcomes.