Computational Social Science

Computational Social Science (CSS) is an interdisciplinary field that uses computational methods to study social phenomena, social structures, and collective behaviour, drawing on tools from computer science, data science, and the social sciences. This course introduces computational techniques for studying societal issues, with a particular focus on large-scale data analytics, social network analysis, and natural language processing.

The course concentrates on issues prevalent in online social media platforms: human behaviour and interactions in online movements (social and political), hate speech detection and mitigation, the impact of misinformation, mental health awareness, and the formation of echo chambers and polarization. Each topic is examined through three lenses: its contemporary relevance, the computational requirements it presents, and the state-of-the-art techniques used to address it. The course also includes a dedicated session on the ethical collection and analysis of data. Tools covered include Gephi and Python libraries for data pre-processing, analysis, and visualization.

The first lecture will be a motivational lecture that focuses on understanding the requirements of this course in today’s context and how it can be helpful to understand and resolve several existing societal issues. Additionally, in our next lectures, we will study different kinds of user interactions and how existing computational social science theories can successfully handle those. From the next lectures, we study a societal problem and a few selected existing research works to resolve that problem. We explore topics, such as, hate speech, anatomy of protests, misinformation, polarization in society and mental health awareness in detail. For each of these issues, we study computational techniques that can handle high volume of data. These techniques can broadly be divided into two techniques. Firstly, techniques related to social network analysis, which will use to study influential individuals or creations of communities.  The second approaches comprise of NLP techniques, such as sentiment analysis, emotion analysis, and topic modelling.

• To introduce the domain of Computational Social Science
• Understand the basics of social computing and its relationship with social media and analytics
• Analyze how social media influences communication, behavior, and societal interactions.
• Discover how social networks and human dynamics create social systems and recognizable patterns
• To introduce the concepts of Social Network Analysis
• To introduce basic concepts of Natural Language Processing

After completing this course, students should be able to

• to work on Social Media Data.
• Create, and manipulate social network data.
• Analyse textual data available from various social media sources.
• Web scrape online data, create a social network visualization with it, and use data science to analyze its content

Network Science by Albert-László Barabási

Info not available