Business Cases with Data Science
Objectives
In business cases, data scientists work with business stakeholders to define the problem, gather data, apply appropriate data analysis methods, and communicate the results to stakeholders. These tasks require combining technical skills and an understanding of the business context.
The general objective of this course is to demonstrate the value of data-driven decision-making by solving specific problems businesses face. Students will be involved in all phases of a business case, including collecting and cleaning data, applying data analysis techniques, and presenting the results and insights in a meaningful and impactful way.
General characterization
Code
200208
Credits
7.5
Responsible teacher
João Carlos Palmela Pinheiro Caldeira
Hours
Weekly - Available soon
Total - Available soon
Teaching language
Portuguese. If there are Erasmus students, classes will be taught in English
Prerequisites
Students should know Statistics, Machine Learning, Data Mining, and Python and have good computer user skills.
Bibliography
Teaching method
Initially, the course is based on theoretical classes (introduction to business cases and the CRISP-DM process model). After that, an example of a case study will be presented, including the demonstration of its source code, interpretation of the results, and examples of decisions to be made based on the results. After that, a case is introduced to students. Students will have two to three weeks to work on the case. During those weeks, students will have Q&A practical and theoretical classes. Then, students will have to make a presentation about the case, submit a report, and all the code developed. The process will be repeated for the four business cases under study.
The work on the case will be done in groups of 5 students each.
Applications used: Microsoft Excel, Python, Jupyter notebook, Microsoft visual code.
Evaluation method
Due to the application-based design of the course, evaluation is continuous and based only on the projects. There is no exam in the course.
All evaluation grades are on a scale of 0-20.
- Group projects:
- The minimum grade is 8.0
- Each project weights 25%
All submissions should be made via Moodle. Submissions after the deadline will be rejected.
Subject matter
LU1. Introduction to business cases
LU2. Introduction to CRoss-Industry Standard Process for Data Mining (CRISP-DM) methodology
LU3. Communication and Data visualization
LU4. Customer segmentation case
LU5. Market basket analysis case
LU6. Demand estimate case
LU7. Customer “churn” forecast case
Programs
Programs where the course is taught: