Introduction to Data Science

Recommended Prerequisites

MI/SP I - pass is not mandatory but attendance is advisable in order to have basic knowledge of descriptive statistics.

Teaching Methods

In the lecture classes, the fundamental concepts are presented and contextualized in real problems. The expositive and dialectical method is used.

In the theoretical-practical classes examples of practical problems are introduced followed by exercises to be solved by the students, using general access computer tools (Orange) or as students of the UC (Microsoft Excel and SPSS), and tutorials are presented for their use in problem solving.

The problems offered to students will be integrated with the contents taught in different curricular units of the 1st semester of MIM.

Learning Outcomes

Be able to:

- Understand the different natures of data and their implications for analysis

- Plan data acquisition, its organization and proper storage

- Transform data in accordance to the needs

- Apply usual data description techniques and graphical presentation

- Recognize the differences between numerical data analysis and text data analysis

- Know some advanced ways of generating and integrating data with the biomedical reality.

Work Placement(s)

No

Syllabus

1.Data acquisition and organization:

Data acquisition in the digital age

-Smart sensors, apps and the internet of things

-Imaging data: the cases of radiography, CT and scanners

-Text data: the cases of papers in medical journals or other biomedical texts

-API’s use to automatically obtain internet data

Automatic data generation using Microsoft Excel

Organization, storage and data protection

-The importance of csv and txt files and its conversion to usual formats

-Data management using Microsoft Excel

2.Visualizing data - extracting knowledge from data

Summary statistics using Microsoft Excel

Data representation and visualization:

-How to create tables automatically – from SPSS to Microsoft Excel

-How to obtain good plots/charts in Microsoft Excel

-Introducing data visualization in Orange

Data and text mining examples applied to biomedical data

-Machine learning

-NLP (Natural language Processing)

Head Lecturer(s)

Bárbara Cecília Bessa dos Santos Oliveiros Paiva

Assessment Methods

Assessment
Project: 100.0%

Bibliography

- Knaflic, C. N. (2015). Storytelling with data: a data visualization guide for business professionals. Hoboken, New Jersey: John Wiley & Sons, Inc.

- Emc Education Services (2015, Data Science and Big Data Analytics: Discovering, Analyzing, Visualizing and Presenting Data

- Bruce, P. C., & Bruce, A. (2017). Practical statistics for data scientists: 50 essential concepts.