Project in Data Science

– Programming and Technology: (Programming, Programming for Data Science, and Databases)
– Analysis and Modelling: (Applied Statistics, Multivariate Statistics, Artificial Intelligence, and Machine Learning)
– Management – Business Context: (Business Management and Organisation, and Management Accounting)

Classes are taught on a theoretical-practical basis, using computers and digital tools. A dynamic teaching and learning process will be used, fundamentally interactive, supported by digital tools and based on the Project-Based Learning (PBD) model. After the constitution of each working group and the choosing of the project theme, the classes will be used to introduce the necessary concepts, methodologies and platforms, as well as to monitor the development of each project.

Course Unit Objectives
– Project Execution: To model and develop a moderate-scale data science project following the Project-Based Learning (PBL) methodology.
– Problem-solving: To identify business needs and utilise data science to find solutions and generate actionable knowledge.
– Knowledge Integration: To consolidate and aggregate concepts, methodologies, and techniques acquired throughout the academic path and current classes.
– Documentation and Dissemination: To produce high-quality technical documentation and define the best strategies for disseminating the results obtained.

Competencies to be Developed
– Project Management: Ability to plan, monitor progress, and meet project milestones.
– Teamwork: Organisation of workgroups and efficient management of available human resources.
– Data Handling: Technical competence to process and transform raw data into relevant information.
– Communication and Soft Skills: Development of transversal skills through progress presentations and the exhibition of results.
– Critical Thinking: Decision-making capacity regarding problem modelling and the relevance of handled data within the business context.

1. Introduction to the concept of data science project.
2. Selection of a business problem as a project theme.
3. Project development.
4. Communication and dissemination of the results.

Assessment can be conducted in two ways:

1) Continuous Assessment: Conducted during term time and concluded on the final day of classes. It comprises the monitoring and submission of 4 (four) milestones of a data science project. These stages correspond to the project phases of Initiation, Exploration, Modelling, and Finalisation, and must be completed in groups of 2–3 members, following a schedule defined by the lecturer. Each stage consists of practical and documentary components, each worth 5 (five) marks, totalling 20. The specifications for each stage, namely technical requirements for data manipulation, models to be applied, and the format of progress presentations, are defined by the lecturer.

2) By Examination: Held during the regulated examination periods, this comprises the completion of 1 (one) individual data science project of moderate scale and 1 (one) oral defence. 
- The individual project involves the full development of a data science solution, integrating problem identification, data manipulation, modelling, and dissemination of results, and is worth 15 (fifteen) marks (out of 20). It must be submitted by 6:00 pm on the day preceding the scheduled exam date.
- The oral defence consists of a presentation and technical discussion of the project to validate the knowledge, methodologies, and techniques applied, and is worth 5 (five) marks (out of 20).

In either of the aforementioned methodologies, all components are mandatory, and there is no minimum grade required for each component to pass. Failure to complete any component will result in a mark of 0 (zero) for that specific element.
The final grade for the course unit is the arithmetic sum of the marks obtained in each component of the chosen assessment methodology.

NAO