COM 450

2024-11-02 Admin 写评论


Hello, if you have any need, please feel free to consult us, this is my wechat: wx91due


COM 450
Fall 2024
Applied Data Analysis for Business
Assignment 3

General Instructions:

This is the third of four assignments you must complete for COM 450. In keeping with the broader objectives of this course, the objective of the assignments are to offer you the opportunity to apply what you have learned to a new context or business problem with an emphasis on communication of the process and interpretation of the results.

The assignments must be completed individually. Please feel free to collaborate but your final submission should reflect your own work.

The assignment is due on at the start of class on Thursday, October 31, 2024. You must submit your Python code and report to the appropriate Brightspace drop box. No hard copy of the report is required.

Assignment Details:

You have been hired by TELUS, a leading telecommunications company, to develop a predictive model to help the company understand and mitigate customer churn.

Customer churn, where customers discontinue their service, poses a significant challenge for TELUS, impacting revenue and growth. By accurately predicting churn,

Telus can implement targeted marketing campaigns aimed at retaining high-risk customers.

You have decided to use a classification tree model to predict churn based on various predictors. A description of the variables is below:
1. CustomerID: Unique identifier for each customer.
2. Age: Age of the customer (in years).
3. Gender: Gender of the customer (Male/Female).
4. Tenure: Number of months the customer has been with the company.
5. MonthlyCharges: Monthly charges billed to the customer (in dollars).
6. TotalCharges: Total charges billed to the customer (in dollars).
7. Contract: Type of contract the customer has (Month-to-month, One year, Two year).
8. PaymentMethod: Method of payment used by the customer (Electronic check, Mailed check, Bank transfer, Credit card).

9. Churn: Whether the customer has churned (Yes/No).

In your report, you must explain how you built and tested the model’s ability to predict new data. You must also interpret the classification tree itself to provide actionable  insights. These insights should help guide TELUS in designing effective marketing strategies to reduce churn and enhance customer retention.

Instructor’s Technical Notes: I would recommend using a relatively large training set (i.e., 80%) and a relatively small testing/validation set (i.e., 20%). It is also very likely that you will get poor classification performance, especially on the testing/validation set. This is largely due to there being a relatively small number of data points and that the data is simulated. The required data set is available on Brightspace. Starter code hasalso been provided.

In the assignment you should:
1. Briefly describe you chosen data set including summary statistics and
appropriate visualizations. This should give the reader an idea of the data set
before you have done any analysis.
2. Carry out a thorough analysis of you chosen data set using a classification tree.
3. Provide a clear and concise description of the results.
4. Clearly state the conclusions of your analysis in the context of the business problem.
Format Details:
The overall structure is flexible however headings are encouraged, and paragraphs are expected. Writing and referencing should be appropriate for an upper-level undergraduate class. Please do not include a title page. Please include a title and your name in the header. Please refer to the two sample reports on Brightspace.

Length: 2 pages maximum inclusive of tables, figures, and references. Hard copysubmissions must be double sided (i.e., one-piece of paper).

Font Size: 11pt
Spacing: 1.5 line spacing
Margins: 1 inch
Font Type: Any conservative font (ex. Arial or Times New Roman)
Citation Style: Any standard citation style is acceptable.
Use of AI:
Please review the Use of Artificial Intelligence section of the course outline. Recall that the use of AI tools, including ChatGPT, IS permitted to generate Python code but NOT permitted when drafting accompanying reports in this course for students who wish to use them.
Keep in mind that you must cite any AI-generated material that informed your work and use quotation marks or other appropriate indicators of quoted material when appropriate. Failure to do this will be considered an academic integrity violation. You must also indicate how AI tools informed your process and the final product, including how you validated any AI-generated citations, which may be invented by the AI.

While academia is still determining how best to cite AI-generated material, in addition to using in-text citations and a corresponding reference list entry, you must:

• Include the full script of your AI “conversation” as an appendix (this can be submitted with your code to the Brightspace drop box), and
• Indicate, through comments, where you have adapted or changed the AI generated code.
Grading Details

Each assignment is worth 16.25% of your final grade. This assignment will be graded according to the following criteria.

Criteria
 Description
 Marks
Data
• A comprehensive description of the data.
• This may include the data source, details on variables, potential relationships between the variables, descriptive/visual analysis.
/5
Problem
 • A clear description of an interesting/insightful problem that is answerable with the selected data.
 /2
Tool Use
 • A clear description of the tool used and a clear discussion of any user decisions made and why.
 /7
Results
 • A clear description of the results both in technical terms and in the context of the problem.
 /5
Conclusions
 • A set of clear conclusions in the context of the data and problem.
 /5
Formatting, Spelling and Grammar, and Citations
• Report meets the basic formatting requirements.
• Writing and references are appropriate for an upper-level undergraduate class.

/1
Total
 /25

发表评论

电子邮件地址不会被公开。 必填项已用*标注