Comparison of Topic Evolution Analysis with Citations

Papers to compare

Items	Structure of Time Evolving Citation Networks	Detecting Topic Evolution in Scientiﬁc Literature: How Can Citations Help?	Comments
Method & Idea	Three "old-fashioned" approaches (mixture model, clustering, unsupervised scoring)	A probablistic graphical model approach (a variant of LDA)	The methods proposed in the first paper are easy to implement and still capture to some extend the nature of citation network, while the second paper incorporates more details of how a topic of a paper should be influenced based on a generative process, which is hard to implement and used in other problems.
Problem	Analyze the structure of large-scale networks (focus on citation networks) that evolve over time	Understand how topics in scientific literature evolve is an interesting and important problem	Basically similar problem and similar research idea.
Data set	Citation network of Supreme Court	Research papers and the citation network from CiteSeerX	The actual size of the citation network used in the first paper is not mentioned, but we can check the page (http://jhfowler.ucsd.edu/judicial.htm) for detail about the citation network of Supreme Court, where we can see the number of cases is 30,288, which is smaller than that in the second paper, which is 650,000.

Q: How much time did you spend reading the (new, non-wikified) paper you summarized?
- A: 1 hour
Q: How much time did you spend reading the old wikified paper?
- A: 20 mins
Q: How much time did you spend reading the summary of the old paper?
- A: 5 mins
Q: How much time did you spend reading background materiel?
- A: 1 hour
Q: Was there a study plan for the old paper?
- A: Yes
Q: if so, did you read any of the items suggested by the study plan? and how much time did you spend with reading them?
- A: 2 of them, 1 hour on average
Cool idea to compare two research flavors of the same idea and similar problem.