General info

Topics

1. AI: a brief introduction to the effective use of generative AI chatbots

2. Web3: blockchain, cryptocurrencies, tokens, decentralized organizations, decentralized education, and more

3. Generative art: old and new forms of generative art

4. The flow of data science: a brief intro/recap of R and the tidyverse approach to data science

5. Network science: centrality and power, similarity and heterogeneity, community detection and clustering, connectivity and resilience, small world, scale-free networks, epidemics on networks, and more

6. Text mining: sentiment analysis, word and document frequencies, n-grams [Not included in the course edition 24/25]

Teaching methodology

We will experiment with DAE - Decentralized Autonomous Education. DAE is an ongoing educational project adopting the philosophy and the applications of Web3 and AI to propose a novel learning model.

Resources

1. AI

2. Web3

3. Generative art

4. The flow of data science: R for Data Science / Dear Data / Old Dear Data

5. Network science: Networks / Networks, crowds and markets / Dear Data

6. Text mining: Text mining with R / Dear Data

Software

• R - a free software environment for statistical computing and graphics

• RStudio - an integrated development environment (IDE) for R (and Python)

• Processing and p5.js

• ChatGPT

R packages

1. tidyverse: analyze and visualize tabular data

2. igraph, tidygraph, and ggraph: analyze and visualize network data

3. tidytext: text mining

Install with:

install.packages(c("tidyverse", "igraph", "tidygraph", "ggraph", "tidytext"))

Data sets

• Cambridge Bitcoin Electricity Consumption Index

• Electricity maps

• Kaggle

• Our World In Data

• Stanford Network Analysis Platform

• The Colorado Index of Complex Networks

• Dune Analytics

Exam

To pass the exam the student needs to complete an individual project of data science and discuss it orally:

• the project is done individually and analyzes a meaningful case study chosen by the candidate

  • the candidate first formulates relevant research questions on the dataset chosen;

  • hence the candidate performs an analysis using methods and tools investigated during the course in order to investigate the posed research questions;

  • finally, the candidate draws conclusions from the performed analysis and finds answers to the research questions

  • the research analysis is documented in an R markdown document (project document)

  • the candidate must also preparate a brief presentation to discuss the project during the exam (project presentation). The format of the presentation is free.

• the full project, i.e., the used dataset, the project document and presentation is archived on a GitHub repository created by the candidate. The candidate must also add the versions of the project document and presentation in PDF or HTML format. The repository link must be emailed to the lecturer at least 48 hours before the exam date

• the project is discussed orally on the day of the exam using a presentation of at most 20 minutes. See the presentation as the tip of the iceberg of your work. During the presentation, the candidate will be asked to show and comment the code of the project

• the candidate is also asked to answer two randomly chosen pre-given general questions about the topics of the course