Scribd
Upload
171 views5 pages

NUS Capstone Project

Uploaded by

akunpost7
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
171 views5 pages

NUS Capstone Project

Uploaded by

akunpost7
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 5

Required Capstone Project

Executing an End-to-End Analytics Process - Template


<Your Name>

Scenario:

You are assisting a credit analyst who wants to gather, organise and analyse data on credit card
transactions and customers in order to identify fraudulent information. Execute an end-to-end analytics
process to accomplish this task.
The tasks below outline the requirements and the step-by-step guide on how to successfully complete this
capstone activity.

Tools:
To complete this project, you need the following database and software:

 Credit_Database.db
 capstone workbook.xlsx
 Task 5.1 Model Data for Clustering
 Task 5.2 Model Data for Classification
 DB Browser (SQLite)
 Power Pivot in Microsoft Excel
 Power BI
 Orange

Note: Click the software to proceed to their official download links.

Assignment Instructions
Task 1: Extract relevant information from a database
Task 1: Using the 'Credit_Database.db' file, find the total number of customers with multiple
cards using a SQL query.
Note: Reviewing Module 2 – Database: Data Source will help you accomplish this task
Estimated Time: 15 minutes
Tools: DB Browser (SQLite)
Submission: Take a screenshot of your work and paste in the space provided below.

Analytics: From Data to Insights Page 1 of 5


Task 2: Organise and shape data by writing SQL queries
Task 2.1: Using the 'Credit_Database.db' file, find the average sales for each customer
segment using a SQL query

Note: Reviewing Module 3 – Database: Data Queries will help you accomplish this task

Estimated Time: 15 minutes


Tools: DB Browser (SQLite)
Submission: Take a screenshot of your work and paste in the space provided below.

Task 2.2: Find the total number of fraudulent transactions and the total fraudulent amount using
a SQL query
Note: Reviewing Module 3 – Database: Data Queries will help you accomplish this task
Estimated Time: 15 minutes
Tools: DB Browser (SQLite)
Submission: Take a screenshot of your work and paste in the space provided below.

az

Task 3: Creating a data model in Power Pivot


Task 3.1: Using 'capstone workbook.xlsx', create a data model with the four tables in Excel
Power Pivot and a calendar table. Show the relationships between these tables.

Analytics: From Data to Insights Page 2 of 5


Add Cust_ID in the TransactionBase table by looking it up from the CustomerBase
table.
Note: Reviewing Module 5 – Date Warehouse: Data Model will help you accomplish this
task
Estimated Time: 60 minutes
Tools: Power Pivot in MS Excel
Submission: Take a screenshot of your work and paste in the space provided below.

Task 3.2: Create a Pivot table with a monthly percentage (MoM %) change in sales. Then,
create a combo chart with sales per month, previous month sales and %MoM spend.
Add slicers to customer segment and customer vintage groups. The chart and the
records in the Pivot table should change as your selection in the slicer values
change.
Note: Reviewing Module 8 – Data Visualisation: Pivot Tables and Charts will help you
accomplish this task
Estimated Time: 60 minutes
Tools: Power Pivot in MS Excel
Submission: Save your output in .xlsx or .csv format with the file name [Capstone_Initial of first
name + last name_Task 3.2]. Upload the Excel file together with this template filled
with your answers.

Task 4: Create data visualisations using tables, charts, and interactive dashboards in Power BI

Task 4: Using ‘capstone workbook.xlsx’, create a dashboard in Power BI as follows –

1. Create 6 cards showing the following metrics-

1) Total number of cards


2) Total number of Customers
3) Total number of Transactions
4) Percentage of fraudulent transactions
5) Monthly average transaction value sales
6) Total sales/total transactions

2. Create a line chart showing Avg Transaction Value by Month.


3. Create a bar chart with the average age for each customer vintage group.
Analytics: From Data to Insights Page 3 of 5
4. Create a matrix/table with rows displaying the quarters, total sales, total
number of transactions, total number of fraud transactions, %age of
fraudulent transactions and average transaction value for each quarter.
5. Link the charts with the customer segment slicer and month dropdown.

The line chart showing average monthly transaction should not change when you
select a month in the month dropdown.
Note: Reviewing Module 9 – Data Visualisation: Power BI I and Module 10 – Data
Visualisation: Power BI II will help you accomplish this task
Estimated Time: 60 minutes
Tools: Power BI
Submission: Save your output in .pbix format with the file name [Capstone_Initial of first name +
last name_Task 4.1]. Upload the Power BI file together with this template filled with
your answers.

Task 5: Apply the clustering and classification technique to classify information


Task 5.1: Using the 'Task 5.1 Model Data for Clustering' file, identify the customer segments
by applying the clustering technique in Orange.
Note: Reviewing Module 11 – Data Mining: Introduction and Clustering will help you
accomplish this task
Estimated Time: 60 minutes
Tools: Orange
Submission: Save your output in .ows format with the file name [Capstone_Initial of first name +
last name_Task 5.1]. Upload the Orange file together with this template filled with
your answers.

Task 5.2: Using the ‘Task 5.2 Model Data for Classification' file, determine the rules to
identify fraudulent transactions.
Note: Reviewing Module 12 – Data Mining: Introduction and Classification will help you
accomplish this task
Estimated Time: 60 minutes
Tools: Orange

Analytics: From Data to Insights Page 4 of 5


Submission: Save your output in .ows format with the file name [Capstone_Initial of first name +
last name_Task 5.2]. Upload the Orange file together with this template filled with
your answers.

Analytics: From Data to Insights Page 5 of 5

You might also like