Assignment Remit
|
Programme Title
|
MSc in Business Analytics
|
|
Module Title
|
Data Analytics and Predictive Modelling
|
|
Module Code
|
38157
|
|
Assignment Title
|
Individual Proposal
|
|
Level
|
7
|
|
Weighting
|
40%
|
|
Hand Out Date
|
14/10/2024 (TBC)
|
|
Deadline Date & Time
|
03.12.2024
|
12 pm
|
|
Feedback Post Date
|
10.01.2025
|
|
Assignment Format
|
Essay
|
|
Assignment Length
|
|
|
Submission Format
|
Online
|
Individual
|
Module Learning Outcomes:
This assignment is designed to assess the following module learning outcomes. Your submission will be marked using the Grading Criteria given in the section below.
LO1. Critically discuss business analytics and related fields and the business analytics project life-cycle application, including determining whether the principles and concepts of business analytics are applied to a business venture.
LO2. Critically evaluate the various life-cycle choices and considerations for a business analytics project, including project scope, data collection, understanding, and analytical modelling.
Assignment:
In this assignment, the students are expected to propose a business problem to investigate and analyse using data analytics and predictive modelling methods. This essay is a proposal written individually by students and is the input for the second assignment. In the individual proposal, the following are expected:
1. Define a business problem (challenge/issue) based on your interest. This problem (challenge/issue) should have the potential to be studied, investigated, and analysed by data. The importance of the problem/topic should be discussed and elaborated. The idea of the business challenge could emerge from news, social media, science, books, articles, your personal interest, experience, working experience, etc. The area considered for study is not limited and could cover any business or industry (manufacturing, service, sport, etc.) at any level (local, regional, national, global, etc.).
2. Determine the research questions and research objectives of your proposal. The research questions/objectives should try to employ data analytics and predictive modelling methods, including (i) clustering algorithms, (ii) classification algorithms, (iii) predictive models, etc. Try to limit the research questions/objectives between 3-5.
2. Extract the features (variables, attributes, criteria, etc.) required to respond to the research questions or achieve the objectives. These features could be based on the current literature or the studied case specifications. Hence, a brief literature review is required to extract the features/factors/variables from reputable and scientific databases.
3. Choose a dataset related to your selected business (secondary data). Ensure that the dataset is free from copyright restrictions. Opt for a dataset that appeals to a specific audience. While you can explore various sources independently, lecture materials, especially from lecture 1, can guide you in finding appropriate datasets (e.g. https://www.kaggle.com). If feasible, you can also extract data from public websites. Aim for a dataset of moderate size at least, comprising over 200 entries (observations, records, rows, or objects) and encompassing more than five different aspects (variables, features, attributes, or columns). Only secondary data is recommended in this module based on time limitations and ethical considerations. Moreover, explain the rationale behind choosing your specific dataset, detailing the nature of the data it contains. The link to your selected data set should be shared in your proposal.
4. The data mining framework, data types and data-gathering procedure should be clarified. Based on the features (variables), it is expected to demonstrate the different data types required to examine each research question. Moreover, the specifications of each data set, such as data level, data type, data time, data structure, etc., are required.
5. The data analysis process/framework should be discussed in detail. Each proposal should illustrate the data analysis framework applied in the second assignment (you can employ online tools such as Whimsical (https://whimsical.com) or Canva (https://www.canva.com)). The data analysis framework is based on the research problem, questions and objectives and is preferred to cover (i) data pre-processing (e.g., data cleaning, data validation, data reduction, etc.), (ii) data processing (e.g., clustering, classification, statistical analysis, predictive modelling, etc.). The selected methods in the data analysis stage should respond to the research questions. Hence, the research questions must be linked to the data analysis methods.
6. Share the expected outcomes of your proposal and the possible limitations. The limitations might arise from the problem, from the data, from the data gathering approach, from the data set, from the data analysis methods, etc.