Temporal Analysis of Terms in Blogs

27 Julho 2011
Sem comentários

Filipe Coelho

4th Doctoral Symposium on Informatics Engineering (DSIE), February 2009, Porto, Portugal


Blogs are becoming extremely popular, revealing the most relevant topics for their social communities on a daily basis. The work presented here has focused on the temporal analysis of terms usage in blogs, specifically the Portuguese emph{SAPO Blogs} collection, to find the most relevant terms occurred during the first half of 2008. The gathered information was stored and processed by means of a data warehouse, which facilitated the necessary calculations for terms analysis by the relevance and interestingness ranking algorithms. Term clouds were used to show the comparison between these algorithms, allowing us to quickly determine that interestingness ranking produced the best results for this collection.

Download Paper as PDF 

Download Presentation





Sem comentários