Week Two: May 24 - May 28

This week I got to work with real dataset. I started off working with the DBLP Computer Science Bibliography. We already have a dataset in our database. My task is to extract data from database and calculate the parameter estimation for the centrality metric. I had a lot of fun working with real dataset. I learned several new techniques in MySQL. It was totally different from learning in the classroom. However, I had struggled with extracting the date of every publication from the dblp dataset due to the complicated schema of dblp. I finally managed to overcome it and get the data ready for building the adjacency matrix for the citation network.

Next step is to create adjacency matrices for the citation network. Each matrix represents a snapshot of the network at a particular time. I chose the UJMP, Universal Java Matrix Package, to help me dealing with large matrix operations in Java. After obtaining the necessary adjacency matrices, I wrote a Java program to calculate the number of chains in the citation network at particular length. The result will be used to estimate alpha which sets the length scale of interactions.

Lessons Learned:
1) Matrix multiplication is not commutative. Don't ever forget it!
2) Graduate students who sit next to you in the office could be really helpful. It takes them two minutes to solve some problems that you have been wrestling with for five hours. However, try your best first, and then ask for help after having too many failed attempts.

Weekend:
Anon took me and other two Thai students to Venice beach on Friday evening. It was nice and pretty crowded. There are a lot of tattoo and piercing shops along the beach. We hanged out and had dinner there. The food was not so good, but the sunset compensated for that.