CS 4704

How do memes spread on blogs? How and when does a hashtag become popular? Can we forecast viral content? How can we harness information cascades to make ’real-world’ predictions? Can Twitter foretell the Flu? In this course, we will develop projects that use web mining, data analytics, computational social science, and applied machine learning to model, analyze, characterize dynamical processes (like information cascades) on social networks. Networks provide enormous potential both to address long-standing scientific questions and also to inform the design of future social computing applications. Social networks, thus pose interesting challenges and questions, which will motivate projects in this capstone course: How is information in a social network created? How does it flow and mutate as it spreads through the network? How do we leverage the information flow to develop real-world predictive applications?

Course Specifics:

The first four-five weeks will comprise of lectures to cover some background necessary to work on problems in this space. After which, we will meet only to learn about some cutting edge research via presentations from CS faculty and senior graduate students and on a need basis for discussing project progress.
This is a 4000 level capstone course, so senior level work is expected. Students should be prepared to apply what they have learned in prior courses (like algorithms, database concepts, etc.) and to hone their skills at learning a new field using primary references as well as some secondary resources.
Students work on team term projects, in groups of size 4-5. Students will be responsible for (with input from the instructor):

Problem formulation
Model development
Implementation & Experiments
Analysis and actionable insights

Final deliverables will be an implementation/working prototype, a report, and a poster to be presented at the Virginia Tech Undergraduate Computer Science Research Symposium (VTURCS) at the end of April. Select projects, based on results, will be communicated to competitive data mining conferences or workshops.

course number	instructor	title
CS 4984	Siddharth Krishnan	Capstone in Social Network Analytics