CIS111-6: Intelligent Systems and Data Mining Assignment Brief: WR Artefact development

CIS111-6: Assignment Brief

What am I required to do in this assignment?

Task

Students will develop an AI solution for saving the cost of a direct marketing campaign by reducing False Positive and False Negative decisions. A cost efficient solution is expected to support the marketing campaign with accurate predictions for a given client profile.
Examples of cost-efficient solutions for direct marketing are provided on the UCI Machine Learning repository describing a Bank Marketing problem.

Examples of previous outstanding assignment reports and MSc projects

1. Bank Marketing by Torubein and Duke (a Springer paper)
2. Stock Market Prediction by Dimitar
3. Stock Market Prediction by Dariusz
4. Stock Market Prediction by Philip (a Springer paper)
5. Medical Imaging by Chris

How students will work

Each student is expected to run individual experiments to find an efficient solution and describe experimental results in an individual report. Students could choose one of three options to work on the assignment: (i) as a group manager, (ii) as a group member, or (iii) as an individual. If students will work in a group, the group manager arranges the comparison and ranking of designed solutions.

Method and Technology

To design a solution, students will use Data Mining techniques such as Decision Trees. Students are recommended to use R scripting (for beginners) via an online RStudio Cloud platform. Other scripting languages such as Python supported e.g. by Google Colab online platform are recommended for advanced students.

Tutorial (to meet the core assignment requirements)
Here is a Python implementation (reproducible in Google Colab) along with a description. This tutorial outlines the steps required to meet the Threshold Expectations described in the section below.

Project Code and Data

Download the Bank Marketing Data (bank-additional.csv) and R script (A1_bank_marketing.R) which are required for individual experiments. Other datasets from Kaggle or UCI, which are related to the unit, could also be used.

Report submission and template

• Each solution will be evaluated in terms of the assignment requirements that are defined for the threshold and advanced levels described below. All submissions are made via BREO as word or pdf files.
• Students are recommended to use a template for reports.
  BREO plagiarism and AI content scores are expected < 20% excluding for codes, generated outputs, and references. Students have multiple attempts to revise and resubmit their reports before the assignment deadline.

Is there a size limit?
5000 words

What do I need to do to pass? (Threshold Expectations from UIF)

1. Follow a RStudio (or Colab) tutorial to create an individual account (10%)
2. Create a new project containing the given script and data set (15%)
3. Apply a Decision Tree technique to solve the assignment task (25%)
   How do I produce high quality work that merits a good grade?
4. Explain how the parameters of an AI technique influence the prediction accuracy (10%)
5. Run experiments to verify the solutions designed on a related benchmark dataset
6. Analyse and compare the results of experiments (10)%
7. Describe the main findings and insights from the carried out experiments (10%)
8. For A grade (>70%) students need to demonstrate excellence of assignment reports at a publishable level, as required by Uni (see examples provided in this brief)

How does assignment relate to what we are doing in scheduled sessions?

Data Mining techniques and use cases developed in R will be considered during lectures and tutorials.

How will my assignment be marked?

Your assignment be marked according to the threshold expectations and the criteria on the following page.
You can use them to evaluate your own work and estimate your grade before you submit.