Chapter

The Influence of Technology on Social Network Analysis and Mining

Volume 6 of the series Lecture Notes in Social Networks pp 275-301

Date:

Cross-Domain Analysis of the Blogosphere for Trend Prediction

  • Patrick SiehndelAffiliated withL3S Research Center, Leibniz University Hannover Email author 
  • , Fabian AbelAffiliated withWeb Information Systems, Delft University of Technology
  • , Ernesto Diaz-AvilesAffiliated withL3S Research Center, Leibniz University Hannover
  • , Nicola HenzeAffiliated withL3S Research Center, Leibniz University Hannover
  • , Daniel KrauseAffiliated withL3S Research Center, Leibniz University Hannover

* Final gross prices may vary according to local VAT.

Get Access

Abstract

In the recent years blogs became an important part of the web. New technologies like smartphones emerged that enable blogging at any time and make blogs more up-to-date than ever before. Due to their high popularity they are a valuable source of information regarding public opinions about all kind of topics. Blog postings that refer to products are of particular interest for companies to adjust marketing campaigns or advertisement. In this article we compare the blogging characteristics of two different domains: the music and the movie domain. We investigate how chatter from the blogosphere can be used to predict the success of products. We analyze and identify typical patterns of blogging behavior around the release of a product, point out methods for extracting features from the blogosphere and show that we can exploit these features to predict the monetary success of movies and music with high accuracy.