Team re-occurrence measures for analyzing style

Sarkar, Avik; De Roeck, Anne and Garthwaite, Paul (2005). Team re-occurrence measures for analyzing style. In: Proceedings of the SIGIR 2005 Workshop on Stylistic Analysis of Text for Information Access. (Argamon, S.; Karlgren, J. and Shanahan, J.G. eds.), ACM Press, pp. 28–36.



In this paper, we propose to investigate style through modeling burstiness in the occurrence patterns of terms in different collections. We set out a fine grained model that looks at gaps between the successive occurrence of the term using a mixture of exponential distributions. A Bayesian framework allows flexibility in fitting the model. The parameter estimates are then studied to understand the distributional properties of a term in various collections. We investigate the behaviour of a range of terms and conclude that the model brings out useful features that may be deployed in the analysis of style.

Viewing alternatives

Item Actions