Big Data Analysis for Social Scientists

Big data involves data on:

(1) people (social web) e.g. online social networks (e.g. Facebook), microblogs (e.g. Twitter);

(2) information (WWW) e.g. web pages, clickstreams;

(3) things (sensor web) e.g. phones, temperature sensors, and

(4) places (geospatial web) e.g. geology, land use maps.

This course is focused on collecting and analysing data from the social web and the WWW.

In this course, you will learn how to:

• Collect data from Twitter, Facebook, YouTube and Web 1.0 websites. Who are the actors, and what actor attributes are available for them?
• Construct, analyse and visualise networks of people and organisations (social networks) and terms (semantic networks). How can we find connections between actors, and how can we use social network analysis to understand the social scientific meaning of such connections?
• Extract and analyse text data. What text can be attributed to these actors, and what does analysis of this text tell us about the actors and society as whole?
• Conduct temporal analysis. How can we study behaviour over time, identifying significant events or trends?
• Identify and engage with advanced techniques for dealing with very large datasets, including software optimisation and sampling techniques.
• Utilise social theory to engage with and reason about the challenges and opportunities of big data, including the interpretation of findings and methodological considerations.

The main software used in the course is R, but we also cover the use of Gephi for advanced visualisation. Data collection will mainly be via an R package for collecting and processing social media data (created at the VOSON Lab specifically for use in this course) and the VOSON software (for collecting WWW hyperlinks and text content). We also provide R scripts covering other important packages for data analysis such as: igraph (network analysis and visualisation), tm (text mining), RTextTools (supervised machine learning for text classification), wordcloud (text word clouds and term frequency visualisation), plyr and stringr (text sentiment analysis), and topicmodels (topic modelling of textual data).

The target audience for this course is people with a fairly strong technical background. The course will be particularly appealing to social scientists who want to become more computationally literate, and those from technical disciplines (e.g. computer science, engineering) who want to become more socially literate.