| 2.9. Pushpa el at.(2012) [14] Link analysis algorithms like Pagerank and HITS are significant extensions to Citation count, for determining the importance of scientific papers in a bibliographic Network. The original Pagerank algorithm is insensitive to the search topic. It attempts to assign high rank to the Web pages based solely on the link structure of the Web. And the traditional Pagerank algorithm doesn’t include temporal dimension in its analysis. We propose a novel approach to include temporal dimension and topic-specific information in Pagerank computation, and study its integrated influence in citation analysis. Our contributions are as follows:
Topic-specific Pagerank are calculated in an unsupervised manner using Latent Dirichlet
Contemporary research papers in a given topic t are located by including temporal
Dimension in the search results.
Authoritative research papers in a given topic t are located by dividing each document’s
Incoming flow variably among its children based on the impact factor of the journal.
Our improved study of the proposed model’s output on High Energy Particle Physics (HEP) dataset, demonstrates its ability to identify contemporary and authoritative documents in a given topic.
Dostları ilə paylaş: | 
