代写MGRC20007 Machine Learning for data-driven decision making 2024-25代做Python编程
- 首页 >> Algorithm 算法Assessments Handbook
MGRC20007 Machine Learning for data-driven decision making
2024-25 – UG
Unit Assessment Breakdown
This unit uses two modes of assessment, explained in detail below:
|
Mode |
% of unit mark |
Brief Description |
Length |
|
Group Assessment |
30% |
1,500-word report where students work in groups to solve a business problem using machine learning. They conduct initial data analysis, develop a preliminary model, and create a comprehensive group report. |
1,500 words |
|
Individual Assessment |
70% |
2,000-word report that builds on the group preliminary work. Each student selects a specific aspect of the project to explore in more depth. This could involve refining the model, addressing limitations, or applying additional analytical techniques to improve the results. |
2,000 words |
Assessment 1 – Group Project
Link to Unit Learning Outcomes
This assessment evaluates the following unit intended learning outcomes (ILO):
• ILO1: Demonstrate an understanding of how decisions are made in
organisations, how sustainable development influences business decision- making, and how business analytics can support decision-making.
• ILO4: Work effectively in a team to develop data-driven recommendations with the potential for a positive impact in practice.
In the Group Assignment, students are divided into teams to tackle a data analytics project aimed at predicting customer churn for a bank. This assignment evaluates specific learning outcomes (ILO 1 and ILO 5) by focusing on how teams apply machine learning techniques to solve a real-world business problem and effectively communicate their findings. Each team member contributes to the collaborative process, from data exploration to model development, ensuring a comprehensive understanding of the project and its implications. Collaboratively, they conceptualise and implement the project, applying data mining techniques effectively.
The deliverables for this assignment include:
• Report (1,500 words): Each team crafts a report similar to a management consultancy document. It outlines the project strategy, encountered challenges, team reflections, and the outcomes achieved.
The group report will be submitted on Turnitin, and further details will be provided closer to the time. The due date for the group report and presentation will be 31st October 2024 13:00 GMT.
30% of the total unit mark
Assessment 2 – Individual Coursework
Link to Unit Learning Outcomes
This assessment covers the following unit learning objectives:
• ILO2: Use mathematical tools to formulate decision-making problems, develop solutions, and provide recommendations based on analytics.
• ILO3: Design and develop suitable predictive analytics solutions to business decision-making problems within the limits of time and resources available.
• ILO4: Implement and evaluate a variety of predictive models and improve on their design to meet business needs and requirements.
Assessment Instructions
The purpose of this assessment is to assess your ability to independently apply and extend the machine learning concepts covered in this unit, specifically addressing ILOs 1 and 5. You are required to submit a 2,000-word report that builds on the group project by refining and enhancing the customer churn prediction model developed in the group work. Your report should demonstrate a deep understanding of the techniques used to improve the model, discuss the ethical considerations involved, and analyse how the refined model can be strategically applied within a real-world banking context. This assignment challenges you to critically evaluate the model effectiveness/accuracy, address potential biases, and propose actionable business insights based on your findings. We encourage the integration of theoretical knowledge with practical application, reflecting real-world business analytics challenges in an individual context.
Important information about the report
The report should be 2,000 words (+/- 10%) in length. Please keep in mind the following points:
• This is not an essay assignment. You should not describe or analyse theory and models in isolation. You will be assessed primarily on how you apply the concepts and models from this unit to a real analytics problem, evaluate its effectiveness, identify and recommend improvements that could be made.
• However, you should include academic references where relevant e.g., the source of models used in your report, supporting data. Please note that you cannot cite the lecture slides! Any quotations from such sources should be properly referenced including page numbers using Harvard referencing style, with full details included in the references section.
• It is recommended that your work should have between 10 to 20 references. Please use Harvard referencing throughout.
• Throughout the report, you will need to focus on both the presentation and clarity of any models, diagrams and tables, as well as the quality of your analysis. Consider how to best present your work to make it look as professional as possible e.g., using a contents page.
• Tables and Figures must be labelled with a caption, “Figure 1: Diagram of … .” etc. with the text referring to the Figure, or table, as Figure 1, etc.
• Reading beyond the course materials is vital.
• The use of graphs and diagrams to illustrate your analysis is strongly encouraged.
• You are reminded that this assignment is about Business Analytics, not data science nor statistics. The mathematical justification and specific number of analyses you applied is not as important as your ability to conduct a clear analysis of the business implications of those data analytics technique.
Report Structure
Please note that the structure below is the recommended format for the assessment.
Recommended Report Structure for the Individual Role-based Assignment
1. Title Page
• Title of the report
• Student ID (not your name)
• Course title
• Date of submission
• Word count
2. Executive summary
• A brief summary of the report purpose, key findings, and conclusions.
3. Introduction
• Overview of the group project and the machine learning model developed. Which model did your group use and why?
• Objectives of the report.
• Brief introduction to the importance of customer churn prediction in the banking sector. Use references to support your arguments.
4. Model improvement
• Detailed description of the enhancements made to the original model (e.g., hyperparameter tuning, feature engineering, cross-validation).
• Justification for the chosen methods and techniques.
• Discussion of the impact of these improvements on model performance, including any new metrics or validation results.
5. Ethical considerations
• Analysis of potential biases in the model (e.g., gender, age, geography) and their implications.
• Discussion on the ethical use of predictive models in banking, particularly in customer retention.
• Proposed strategies to mitigate identified biases and ensure ethical application of the model.
6. Strategic application
• Application of the refined model to a specific business scenario within the bank (e.g., targeting high-risk customers for retention or SMEs).
• Analysis of how the model predictions can inform. strategic decisions.
• Discussion of the potential business impact of these decisions, supported by data from the model.
7. Critical reflection
• Personal reflection on the learning process and the challenges faced during the project.
• Evaluation of the strengths and weaknesses of the model and the overall approach.
• Suggestions for further improvements or alternative approaches.
8. Conclusion
• Summary of key findings and insights from the report.
• Reiteration of the importance of model refinement, ethical considerations, and strategic application in business analytics.
9. References
10.Appendices (if applicable)
• Additional data, charts, Python scripts, or other material referenced in the report but not included in the main body.
You should also make sure that you are fully aware of the School's policy on plagiarism. You should be aware that you cannot later claim that you did not know the rules and regulations. Copying material from similar essays that can be found on essay websites is not acceptable and can lead to disciplinary action. See
https://www.bristol.ac.uk/students/support/academic-advice/academic- integrity/plagiarism/ for full information.
Various websites claim that they help students by showing them what is expected on a typical assignment such as in Business Analytics. These tacitly encourage plagiarism and copying, which does not demonstrate true understanding. It is more important to develop your own voice and your own abilities in writing and research, and to show that you can see how operations function in any real-world setting. All assignments are scanned via plagiarism detection software
At the University of Bristol, while recognising the potential of AI technologies in enhancing learning, the emphasis is on ethical usage and academic integrity. Students are expected to use AI tools, like ChatGPT, responsibly, adhering to the university's guidelines that discourage dependency on such technologies for completing academic tasks. For detailed information on the correct use of AI in your studies, including the university's policies and recommendations, please refer to the Using AI at University guideprovided by the Study Skills team.
Each student must submit their completed portfolio assignment electronically via Canvas. Assessment due date is 28th November 2024 at 13:00 (UK time).