Task 1: Topic Approval
Contents
Choosing a Topic
Examples
Submitting the Approval Form
Restricted Information Authorization Form
FAQ
Questions, comments, or suggestions?
Warning
Submission process update. Task 1 is no longer signed and
returned by your course instructor. Fill out the approval form and
submit it directly to Assessments as your Task 1. Your course
instructor (or, in the future, an evaluator) will either pass your
submission or return it with comments for revision. This mirrors how
you submit Tasks 2 and 3.
Choosing a Topic
The approval form confirms a viable topic before you invest time and effort
into task 2. A topic is approved when it has the following:
Back to top
1. An application of machine learning (ML).
2. An organizational need for your ML application to help solve.
, 3. A basic outline of your implementation plan.
The “organizational need” requirement gives your project a purpose and
audience. From an assessment perspective, it has little other value, so you
need not worry about your project’s profitability or practical impact. Such
criteria are not assessed. This is a computer science project, not a business or
software project.
Examples
Topic Summary Examples
Example 1 Example 2 Example 3 Example 4
Data: Petal dimensions and species of Iris Samples.
ML application (non-descriptive method): Classify an Iris’s species
using its petal dimensions to train a supervised classification model,
Print to PDF
say SVN or Logistic regression.
Descriptive method: Histograms and barplots showing distributions
of different flower features and confusion matrix illustrating the
accuracy of the classification model.
Organizational Need: Grow Fast Fertilizer Inc. needs to help
customers identify their flowers.
Tasks 1 & 2 Examples
These passing examples have been provided by the evalaution team to help
students better understand what’s expected. They are on-par with average
passing tasks. Use them as a guideline for what evaluators accept in fulfillment
, of the requirements. To best represent what might be accepted, we’ve made
no corrections.
Task 1 Examples Task 2 Examples
Example 1
Example 2
Example 3
WGU Capstone Excellence Archive
The Capstone Excellence Archive includes a wide range of completed projects.
Reviewing them may help get ideas, provide inspiration, and understand the
requirements. However, keep in mind that they all are above and beyond the
requirements. So don’t use these as examples of what’s needed to meet the
requirements. For a more down-to-earth example of what’s required, see the
examples above.
Warning
Some examples in the archives may follow the old template. Please
follow the templates found on this website.
Data
ML applies an algorithm to data, and you can’t apply an algorithm to data -
without data. So finding data should be an early step.
A machine learning application is the hard requirement of the capstone, but
you can’t apply an algorithm to data -without data. Moreover, most datasets
provide a question to answer, e.g., predictions, classifications, or
recommendations for which an “organizational need” can always be found.
Kaggle.com
Contents
Choosing a Topic
Examples
Submitting the Approval Form
Restricted Information Authorization Form
FAQ
Questions, comments, or suggestions?
Warning
Submission process update. Task 1 is no longer signed and
returned by your course instructor. Fill out the approval form and
submit it directly to Assessments as your Task 1. Your course
instructor (or, in the future, an evaluator) will either pass your
submission or return it with comments for revision. This mirrors how
you submit Tasks 2 and 3.
Choosing a Topic
The approval form confirms a viable topic before you invest time and effort
into task 2. A topic is approved when it has the following:
Back to top
1. An application of machine learning (ML).
2. An organizational need for your ML application to help solve.
, 3. A basic outline of your implementation plan.
The “organizational need” requirement gives your project a purpose and
audience. From an assessment perspective, it has little other value, so you
need not worry about your project’s profitability or practical impact. Such
criteria are not assessed. This is a computer science project, not a business or
software project.
Examples
Topic Summary Examples
Example 1 Example 2 Example 3 Example 4
Data: Petal dimensions and species of Iris Samples.
ML application (non-descriptive method): Classify an Iris’s species
using its petal dimensions to train a supervised classification model,
Print to PDF
say SVN or Logistic regression.
Descriptive method: Histograms and barplots showing distributions
of different flower features and confusion matrix illustrating the
accuracy of the classification model.
Organizational Need: Grow Fast Fertilizer Inc. needs to help
customers identify their flowers.
Tasks 1 & 2 Examples
These passing examples have been provided by the evalaution team to help
students better understand what’s expected. They are on-par with average
passing tasks. Use them as a guideline for what evaluators accept in fulfillment
, of the requirements. To best represent what might be accepted, we’ve made
no corrections.
Task 1 Examples Task 2 Examples
Example 1
Example 2
Example 3
WGU Capstone Excellence Archive
The Capstone Excellence Archive includes a wide range of completed projects.
Reviewing them may help get ideas, provide inspiration, and understand the
requirements. However, keep in mind that they all are above and beyond the
requirements. So don’t use these as examples of what’s needed to meet the
requirements. For a more down-to-earth example of what’s required, see the
examples above.
Warning
Some examples in the archives may follow the old template. Please
follow the templates found on this website.
Data
ML applies an algorithm to data, and you can’t apply an algorithm to data -
without data. So finding data should be an early step.
A machine learning application is the hard requirement of the capstone, but
you can’t apply an algorithm to data -without data. Moreover, most datasets
provide a question to answer, e.g., predictions, classifications, or
recommendations for which an “organizational need” can always be found.
Kaggle.com
