Applied Business Intelligence – Assignment 1

Module Name: Applied Business Intelligence
Module Code: 55-600268
Assignment Title: Assignment 1
Submission Type: Group Presentation

1.1 Learning Outcomes

This assignment assesses the following learning outcomes:

1. Describe and critically evaluate the role and relevance of business intelligence and analytical investigation in solving business information problems.
   1. Explain key concepts underpinning business intelligence, including established theories and current developments.
   2. Apply concepts and justify decisions when modelling and designing practical applications using industry-standard software.

1.2 Assessment Criteria

Assignment 1 contributes 40% of the final module grade. It is a 10-minute group presentation based on analysis of the dataset described later. Each question in Section 1.9 has a specific mark allocation.

1.3 Submission Details

One group member must upload the final PowerPoint presentation to the “001 Group Presentation – Submission Point” on Blackboard by the deadline.

1.4 Presentation Requirements

You are presenting to a bank manager who has no technical background. Therefore:

• Your slides must show clear, easy-to-interpret outputs.
• Avoid SAS code, configuration settings, or technical jargon.
• Explain results in business-focused language.

Each group member must speak. The presentation will be stopped at 10 minutes, followed by 3 minutes of questions.

1.6 Problem Outline

The dataset for this assignment contains five years of financial transactions and customer information from a Czech bank. Your task is to analyse behavioural patterns and identify meaningful account characteristics.

1.7 Data Provided

The SAS dataset is:

czechbk15.sas7bdat

Location on the SHU server:

E:SHUUsers!SharedDataRichABI2223

You must create a SAS library pointing to this directory.

1.8 Query Structure and Variables

The selected variables represent five years of credit activity and multiple types of withdrawals.

Credits

• credit – total money paid into the account
• creditn – number of credit events

Withdrawals

Each withdrawal method includes two variables:

Additional Account Attributes

• account id
• age of primary account holder
• credit card (y/n)
• days account has been open
• loan (y/n)
• second account holder present (1/0)
• sex (M/F)
• frequency of bank statements (monthly, weekly, after-transaction)

1.9 Analysis Requirements

Step 1 — Explore Interval Variables (8 marks)

You must produce summary measures and plots for all interval variables. Interpret patterns, anomalies, and outliers. Use Enterprise Miner’s Explore node to support your investigation.

Step 2 — Explore Binary & Nominal Variables (5 marks)

Generate appropriate plots and fully discuss behavioural patterns.

Step 3 — Transform Variables for Clustering

Use the Transform node → Maximum Normal method.

a) Interpret transformations (2 marks)

Explain which variables were transformed, why certain variables remained unchanged, and include evidence (e.g., SAS transformation table).

b) Evaluate transformation success (8 marks)

• Show before/after plots.
• State which version (original or transformed) is better for clustering.
• Produce your final list of variables for cluster analysis.

Decision Tree Analysis – Predicting Second Account Holders

a) Interpret Decision Tree & Fit Statistics (6 marks)

Explain the segmentation logic and how the model performs.

b) Apply Model to a New Customer (3 marks)

Using the provided customer values, determine likelihood of having a second account.

c) State Reservations About Using the Model (3 marks)

Discuss issues such as bias, data imbalance, overfitting, missing values, etc.

2.0 Presentation Requirements

Your PowerPoint must be clean, professional, and structured logically. The group presents for a maximum of 10 minutes, with 3 minutes for questions. Marks will also be awarded for your ability to answer questions effectively.

Total Marks Available: 40