Pagerank algorithm implementation using Hadoop MapReduce and Java language.
- According to the transition matrix input(transition.txt) build a relationship model
- Calculate the weight or transiton fact between pages
- PageRank1 = Transition X PageRank0
- Sum up each unit weight to get new rank model
- Converge above steps N times
$ hadoop jar pagerank.Driver -trans /transition -rank /pagerank -unit /output -times 5
usage: pagerank.Driver
- -rank input rank file dir
- -times times of convergence
- -trans input transition file dir
- -unit unit output dir