Computational Social Science Capstone Project (Coursera)

Overview

CONGRATULATIONS! Not only did you accomplish to finish our intellectual tour de force, but, by now, you also already have all the required skills to execute a comprehensive multi-method workflow of computational social science.

We will put these skills to work in this final integrative lab, where we are bringing it all together. We scrape data from a social media site (drawing on the skills obtained in the 1st course of this specialization). We then analyze the collected data by visualizing the resulting networks (building on the skills obtained in the 3rd course). We analyze some key aspects of it in depth, using machine learning-powered natural language processing (putting to work the insights obtained during the 2nd course).

Finally, we use a computer simulation model to explore possible generative mechanism and scrutinize aspects that we did not find in our empirical reality, but that help us to improve this aspect of society (drawing on the skills obtained during the 4th course of this specialization). The result is the first glimpse at a new way of doing social science in a digital age: computational social science. Congratulations! Having done all of this yourself, you can consider yourself a fledgling computational social scientist!

Overview

WEEK 1 – Getting Started

For this milestone, you will again web scrape videos from two YouTube channels. You will be assigned two channels to scrape. In contrast to the previous version of this exercise, you will NOT scrape the featured videos of the specified news channel, but the search results of the name of the news channel in combination with your name.

WEEK 2  – Social Network Analysis

In this milestone, you will analyze a social network with help of the software Gephi.

WEEK 3  – Natural Language Processing

In this milestone of our Integrative Lab, you will select two of the key videos identified with help of our SNA, and analyze the sentiment and emotions contained in the comment sections of the videos. We use NLP from IBM Watson for this.

WEEK 4  – Agent-Based Computer Simulations

In this milestone, you will take all the data you created in the previous milestones and use a two-step flow model and discover how ideas can diffuse into society. Through this exercise you will grow your own artificial society from the bottom up.

Teacher

• Martin Hilbert