Skip to content

Home

Jump to bottom
Social Complexity Lab edited this page Aug 31, 2022 · 52 revisions

Intro

Welcome to the wiki for the course Social graphs and interactions (02805) offered by the Technical University of Denmark. This is the main page, where you can access the weekly exercises. If you take a look in the side-bar, you can read about the administrative details (including a very useful course overview), assignments, books, and more.

The class is taught flipped classroom style, where the lecture and homework elements of a course are reversed. You'll be able to view short video lectures before (or during) the class session, so in-class time can be devoted to exercises, projects, or discussions. Check out the Before week 1 lecture to learn more.

IMPORTANT COVID-19 INFORMATION

Lectures

  • Before week 1. Take a look at this page before you do anything. This class most likely works a little bit differently from other classes you've taken. The notebook explains pretty much everything - the rest will be explained during the lectures.

  • Python BootCamp. Python is the key tool we use in this class. If you don't feel 100% ready this notebook offers a quick refresher course.

  • Week 1: Introduction. This week is all about getting started. It's a light load, since we want everyone to get a good start, especially if you're not a Python Ninja, just yet. Thus, there's room for prep, making sure you're all on top of Python, etc. But we also get started on the Network science with an introductory lecture, and playing with NetworkX, the Python library for network analysis. In case the link below doesn't work, you can also see the file here on github, but the videos won't display properly.

    • Reading: Network Science book, Chapter 1. You can get the whole book for free here. Or buy it at the Campus Bookstore.

  • Week 2: Networks I. It's time to learn a more about networks. I have to admit that I love networks. I could talk about them for hours. And that's actually also what I'll be doing for today's lecture. Lots of info from me + some reading for you guys. I'll answer some important questions, such as "Why would anyone care about networks" and "How can you use Python to study networks". In case the link above doesn't work, you can also see the file here on github.

    • Reading: Reading. Chapter 2, and 3 (section 3.1-3.7 ... the most important part is 3.1-3.4, so focus on that) of Network Science. You can find the entire book online here.

  • Week 3: Networks II. This week, we go deeper with networks. Now that you're on top of the basics, we'll look at the research for this week's topic. We'll focus on two key research results from the turn of the millenium that truly kick-started a revolution in our understanding of networks. Specifically, we will look at problems with random networks as models for real networks and how that leads to the Watts-Strogatz model. Then we will discuss scale-free networks and the Barabasi-Albert model. As always, in case the link above doesn't work, you can also see the file here on github.

    • Reading: We will continue with Network Science. We start with the rest of chapter 3 with an emphasis on 3.8-3.9. Then we read Chapter 4, section 4.1 - 4.7 and Chapter 5, section 5.1 - 5.5.

  • Week 4: Networks III, Revenge of the Data Scientist. Today you will be getting your very own dataset from Wikipedia. Working with real data is a pain in the a**, and today you will experience this fact first hand. But you should, of course, be thanking me for providing this experience, since this is what the real world feels like. After your life at DTU, no one will be giving you a nice, cleaned dataset that you can easily load into your favourite data structure. So I hope this experience will be valuable for you as you move through life after DTU. And I promise that you will never fear raw data again. As always, in case the link above doesn't work, you can also see the file here on github.

    • Reading: There's no reading today. You'll have enough to do with running regular expressions and other fun things. So if you're bored, this is a good time to catch up on the stuff from the previous weeks that you haven't looked at.

  • Week 5: Networks IV, Advanced measures. You've done a lot of work retrieving the Marvel/DC networks from Wikipedia. Today the goal is to analyze it and learn something about both network science, comics and Wikipedia itself along the way. As always, in case the link above doesn't work, you can also see the file here on github.

    • Reading: This week, the reading is mostly for reference. It's for you to have a place to go, if you want more detailed information about the topics that I cover in the video lectures. Thus, I recommend you check out Chapter 9 of the network science book. In particular, we'll delve into section 9.4 in the exercises. We will also talk a little bit about degree correlations - you can read about those in Chapter 7.

  • Week 6: NLTK I, Getting started with NLTK. Ok. So we're changing gears. We've looked at the networks in Wikipedia. Now we'll put together the tools for working with the text. This first week is going to be a walk in the park (still feel bad for making you download all those comic book character pages) - so we'll just be installing the software, reading the book a bit, and solving some exercises. Easy-peasy. (Plus, if you're behind, today's light load makes it a nice day to catch up on everything else). As always, in case the link above doesn't work, you can also see the file here on github.

    • Reading: Natural Language Processing with Python (NLPP) Chapter 1, Sections 1.1, 1.2, 1.3. (It's free online) and NLPP Chapter 2.1 - 2.4.

  • Week 7: NLTK II, A mixed bag of useful tricks. Now, let's get real and work with some language/text. I'm taking you through the dreaded chapter 3 of NLPP, talking about TF-IDF as a way to summarize what is important about a document, and we'll be getting into sentiment.

  • Week 8: NLTK III, COVID-19 Comics. We've reached the end of the lectures. And it's a good one! We start by finishing up the work that focuses on understanding the communities via their content (using TF-IDF based word-clouds and sentiment analysis). Then we move on and study epidemic-spreading on the super-hero network in some brand new COVID-ear exercises.

    • Reading: Network Science, Chapter 10.1, 10.2. (This is all you have to read, but I recommend the rest of Chapter 10 if you're interested in epidemics and networks.)

This class has been hand crafted for you by Sune Lehmann

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Clone this wiki locally