Evaluation

Overview

Introduction

This has been a phased approach with each new phase building on the previous effort. While some previous information will obviously contribute to this work, most of this phase will be new work. At this point, the project must be through the modeling stage. As you bring together the work from Phases I and II and evaluate the results, you will most likely want to revisit, alter, or otherwise tune your original hypothesis. You may find some new data, change your descriptor or predictor variables, or make other changes to see if you can add a greater level of confidence to your effort. While engaging in changes to your project, keep an accurate account of what is done so that future efforts will have the benefit of this historical knowledge. Your analysis of the data must be presented in a manner that would be appropriate for repeatability by a fellow data miner. Notes containing information such as outlining your decision processes, thoughts about your analysis, and past test results can be valuable information for any future work.

Background

As you come to the evaluation phase, you will place a greater focus on the examination of the model under the lens of the business problem. Earlier, you reviewed the data for its accuracy and relevancy to the model, but here your interest is in whether or not the totality of the model provides significant value to the business question. You may need to determine the viability of implementation, cost to the company (time, money, oversight), etc. Be aware of having used data that may not be available outside the project engagement. In other words, if deployed, is the data available that is needed to ensure successful operation? Remember, data visualization can go a long way in helping communicate a key concept, issue, complex relationship, etc.

Instructions

Requirements

• Using the CRISP-DM 1.0 and Bui (2020) textbooks as guidelines, submit a comprehensive report augmenting the prior project phases. This must address things such as findings, analysis, next steps, and possible obstacles and/or opportunities in the future. This assignment must be at least 5 pages and include at least 5 scholarly sources other than the course textbooks.
• Use current APA style guidelines.
• Utilize at least 3 images to augment your report (e.g., findings, methodology, sample representation, distribution, relationship, etc.).
• Follow the CRISP-DM framework (helpful with establishing the paper sections, steps, issues to address, etc.).
• Remember your audience and write accordingly.

Primary References

CRISP-DM 1.0. (1999). Retrieved from https://www.the-modeling-agency.com/crisp-dm.pdf.

Bui, Y. N. (2020). How to Write a Master’s Thesis (3rd ed.). Los Angeles, CA: SAGE Publications. ISBN: 9781506336091

Sample Secondary References

Domain data	Internet	Journal articles
Public data sites	Work-related SMEs	Scientific method
Textbook and DVD tutorials	Work-related data

Submit

• Upload your assignment as: Phase_III_YourloginName (e.g., Phase_III_JJSmith).

Note: Your assignment will be checked for originality via the Turnitin plagiarism tool.