2610 - BUSINESS ANALYTICS SPECIAL PROJECT
| Site: | moodle@NovaSBE |
| Course: | 2610-Business Analytics Special Project-2425_S2 |
| Book: | 2610 - BUSINESS ANALYTICS SPECIAL PROJECT |
| Printed by: | Guest user |
| Date: | Saturday, 7 June 2025, 2:13 PM |
1. Course Unit Aims
Business Analytics Special Project can be seen as a capstone-like experience in which the student would demonstrate a broad knowledge of Business Analytics by undertaking hands-on projects with realistic data.
Its main goal is to provide students an effective and meaningful understanding of the business analytics process and function through the concept of self-directed discovery and experiential learning. The primary activity in this course is a semester-wide project intended to produce a business analytics solution which provides substantive business utility, and to demonstrate its technical and economic feasibility in the contextual domain it was developed for.
Upon completion of the several components of the projects, students will demonstrate a broad knowledge and clear understanding of critical concepts, practices and issues in Business Analytics project development and completion.Additionally, students will acquire experience in understanding the organization they are working with, the managerial problems the institution is facing, they will learn how to use technology and leverage data to solve those problems, identifying and understanding all the constraints of its implementation.2. Course Unit Content/Calendar
- Components of the Project:
o Project Charter: a formal document that describes the project in its entirety, including, for instance, the definition of the problem, the goals of the project, the team and their specific roles, the identification of the deliverables, a list of the risks identified and mitigation actions, the data that will be used, among other.
The Data Dictionary, a document that allows anyone to easily understand a dataset, should also be included in this deliverable.
o Exploratory Data Analysis: process of performing initial investigations on data - discover patterns, spot anomalies, test hypothesis and check assumptions with the help of summary statistics and graphical representations.
o Modeling Roadmap: a document specifying the modelling dataset, the models to be trained, and the models' validation and evaluation. It should also describe the Baseline model that will be used to establish a benchmark against which the models will be evaluated.
o Modeling Report: a report in which the results obtained in the modelling stage are described, including results from a bias audit, if applicable, to exploring potential unintended consequences of deploying the model. In the report you should also describe the evaluation metrics used and why, the evaluation of the models tested and the final selection. Moreover, it should also include a section Plan for Testing that specifies how the model adoption and realworld performance may be tested.
o Pipeline: the sequence of steps that need to be taken from raw data to model results. The pipeline is the whole code used, which includes the preprocessing code, the modeling code, and any other code used to produce the results. The output of this component should be a Github page for the Project.
- Regular meetings with the Team’s supervisors and also with the project partner
3. Bibliography
- The main references will depend on the project’s topics. So, they will be provided by each team’s mentors as the project develops.
4. Assessment
A student’s final grade will depend on the quality of the following project components (C’s) (the group mark):
|
o C1 - Project Charter (including Data Dictionary) (20%) |
o C4 – Modelling Report (including the Plan for Testing) (35%) |
|
o C2 - Exploratory Data Analysis (25%) o C3 – Modelling Roadmap and Baseline Model (15%) |
o C5 – Pipeline Completed (5%) |
And also on his/her team’s peer assessment (the contribution mark). The contribution mark is used to weight the group mark, so that a student who makes an average contribution to the project is awarded the group mark, and those who make greater (or lesser) contributions receive more or less than the group mark.
To ensure fairness in the results, a moderation will be conducted before computing the final value of the contribution mark. In particular, the course instructor will
- check that the comments justify the scores given (scores that are not justified can be removed and the average contribution mark will be recalculated).
- check if there are common themes with comments from different group members. oinclude the feedback from the team’s mentors.
Therefore, the final grade will be calculated using the formula:
final grade = (group mark) x (individual contribution mark)/(average contribution mark)
Deadlines:
o Component C1 is due on December, 7.
o Component C2 should be delivered by February, 1.
o The Modelling Roadmap and the first results obtained by the Baseline Model (Component C3) should be presented on March, 6.
o By the end of the 2nd semester (May, 29), the Technical Report (Component C4) must be concluded.
o Finally, the last component C5, as well as the peer assessment should be uploaded on the course moodle’s page on June, 8.