Data Science Semester 221- Assignment 1
Information
• Due Date – Sunday 06/03/2022 11:59pm
• Percentage of Grade – 10%
Copyright By PowCoder代写 加微信 powcoder
• Data required for this assignment is on iLearn in the Assignment 1 folder
• You are required to submit your solutions to this assignment via iLearn. Submit a zipped
folder, containing two knitted PDF or Word versions of your RMD file.
o One file should represent your report to management.
o One file should represent the same report but with all code shown.
• Ensure you adhere to the Academic Integrity Guidelines for Coding.
Task Description
Congratulations, you are a newly employed Data Scientist at Bond Bank. Bond Bank services a wide range of customers and is starting to investigate whether data science can improve business operations. You are the first employee in the Data Scientist role, and there are high hopes for you.
For your first project, Bond Bank wants to improve their marketing campaigns. They have been calling many of their clients to determine whether they would like to subscribe to a term deposit. They don’t currently have a systematic approach for choosing who to call. Phone calls to clients who aren’t interested result in a poorer client experience and wasted time by the caller.
The first question you have been asked by management is: “Is there a way to determine in advance which clients will subscribe to a term deposit as a result of a marketing phone call?”.
To answer this question, you have been provided information about past marketing calls1. This is contained in the “BankMarketing.csv” file on iLearn. The “Assignment Data Dictionary.xlsx” file contains metadata.
Your goal is to create a decision tree which can predict class membership of the Y variable. The marketing team will use your tree’s rules when deciding which clients to contact. The better the predictions, the more cost-effective the marketing team can be. Management also wants you to report how well the tree performs and show some example predictions.
1 Data obtained from https://archive.ics.uci.edu/ml/datasets/bank+marketing
Deliverables and Tips
Deliverables
Your final deliverables will be 2 PDF or Word files, both produced by the same .Rmd script (with different code chunk options). You must submit:
– Technical Report – all code and results shown (like you would share with a colleague on the Data Science team)
– Management Report – only show those things necessary to help support management decision making (this is the one you send to management!)
Use of Models
– You may build more than one tree and compare them, whether that means variations of trees from a single package, or trees from multiple packages.
o If you build more than one tree, you need to compare them and select a final tree to be used.
o Remember that the goal is to develop a final model, not to step management through an extensive comparison of models.
– You will need to show how well the tree performs
– You will need to show some example predictions.
The Management Report
– Focus on good document structure from the beginning – how are you structuring the information in your report?
– Remember that management may not really understand data science, so you will need to find a trade-off between understandability and verbosity. As the data scientist needs to bridge technical and business areas, communication is hugely important.
o Visualisations are excellent for communication
– I expect the management report to be under two pages. Verbose assignments will be
penalised.
The Coding
– Everything required for the assignment is discussed in lectures and workshops. You are welcome to incorporate ideas outside of the delivered content, but it is not required.
– You’re going to use your final model in a later project – make sure you can recreate it.
程序代写 CS代考 加微信: powcoder QQ: 1823890830 Email: powcoder@163.com