Introduction to data science

Course objectives

This course introduces the students to five key facets of a data-based research, where data are obtained by observation:
(1) data wrangling, cleaning, and sampling to obtain a suitable data set,
(2) data management to facilitate efficient access to big data,
(3) exploratory data analysis to generate hypotheses and intuition,
(4) prediction based on statistical methods such as regression and classification, and
(5) communication of results through visualization, stories, and interpretable summaries.

Enrolment requirements and/or entry competences required for the course

-

Learning outcomes at the level of the programme to which the course contributes

  • Apply theoretical knowledge of the fundamentals of the six core disciplines and their relationship within cognitive science.
  • Critically evaluate cognitive science findings and synthesize information to be employed in a collaborative professional environment.
  • Participate in data-driven innovation projects and apply appropriate data science tools.
  • Apply AI tools in concrete tasks and practical contexts.

Course content (syllabus)

  1. Course administration. Overview of the data science field. Supporting technologies for data science. Auditory exercises: introduction to Pandas.
  2. Data handling, where date are obtained by observation: data acquisition, data models, common dataset issues, data reshaping, data cleanup. Auditory exercises: data handling and feature engineering in Python. Project: studying suggested scientific papers.
  3. Data visualization: various graphs for dataset visualization, best practice for data visualization, visualization for special purposes, visualization tools. Auditory exercises: data visualization in Python. Project: studying interdisciplinary scientific papers, selecting one for replication of results.
  4. First view of data: descriptive and inferential statistics. Auditory exercises: descriptive statistics in Python. Project: consultation with assistant regarding the selected scientific paper.
  5. Data annotations and metrics. Auditory exercises: data annotations and metrics. Project: work on replicating the results.
  6. Data acquisition through research: types of research studies and data acquisition methods. Project: work on replicating the results.
  7. Applied linear regression in descriptive data analysis. Data transformation. Linear regression assumptions. Auditory exercises: introduction to regression analysis. Project: work on replicating the results
  8. Midterm exam
  9. Applied supervised machine learning: classification and prediction. Auditory exercises: applied supervised machine learning in Python. Project: finishing work on replicating the results.
  10. Applied unsupervised machine learning: clustering. Auditory exercises: applied unsupervised machine learning in Python. Project: forming a team to improve the results of scientific paper, consultations with assistant.
  11. Introduction to deep learning (neural networks, loss, invariance and equivariance, convolutional neural networks, recurrent networks). Auditory exercises: deep learning in Python. Project: team work on improving the results.
  12. Text handling (text, feature vectors, bag of words, tokenisation, stop words, n-grams, TF/IDF, attention). Auditory exercises: working with text data in Python. Project: team work on improving the results.
  13. Handling graphs and networks (nodes and edges, directed and undirected graphs, centrality measures, Graph convolutional networks). Auditory exercises: working with graph data in Python. Project: team work on improving the results.
  14. Project presentations.
  15. Final exam

Student responsibilities

Lectures, auditory exercises, team work, project work

Required literature

  • Jacob T. Vanderplas, Jake VanderPlas (2016.), Python Data Science Handbook, O'Reilly Media, Inc.

Optional literature

  • Matt Harrison, Theodore Petrou (2020.), Pandas 1.x Cookbook, Packt Publishing Ltd.
  • Alice Zheng, Amanda Casari (2018.), Feature Engineering for Machine Learning, O'Reilly Media, Inc.
  • John D. Kelleher, Brendan Tierney (2018.), Data Science, The MIT Press.