This document describes the process of calculating weighted page rank on a dataset containing paper citations. It discusses:
1) Converting the semi-structured data into a graph by processing the indexes and references.
2) Calculating in-degree distribution and using it to determine weights based on in-links and out-links for each node.
3) Calculating page rank through an iterative process of updating ranks based on the weighted page rank formula over 10 iterations, and joining the results with the in-degree distribution to find the top 10 most influential papers.