Reference no: EM133248024
Assignment:
Use a decision tree classier as the chosen machine learning model you want to build and review the following: I have others in mind but I need a different view.
1. Explain the type and sources of data that will be used. Where does the data come from? What does it represent? Is it quantitative? Is it text? or any other? And how big?
2. Describe the specific requirements for processing and analysis. Is there a need to use a specific analytical model? What is the nature of the expected end results: prediction, pattern identification, classification, decision, etc.? What assumptions need to be verified? How will the results be validated?
3. Explain all performance constraints. Does the analysis need to be completed within a specific time frame? What performance metrics are used? Why are they important?
4. Portray how the product matches the analytical objectives. How do you ensure that the analytical process will indeed provide the type of results expected? How do you measure the extent of the usefulness of the analysis results?
5. Outline the type of visualization necessary. What exactly needs to be visualized? What visualization metaphors/paradigms will be used? What would the visualization add to the analytical process that could not otherwise be conveyed?
6. Explain how you will ensure usability. If one makes an error, will there be an alert? What aspects of the user interface will ensure that the data product is easy to use and intuitive? Will there be a user guide?
7. Describe the interactivity and deployment venues. How will the data product be accessed? Will it be deployed on a cloud platform? A personal website? What can the user do? Can the user choose among different analytical steps? Are there editable parameters for analysis or visual displays?