Regression-Based Summarization of Email Conversations

In this paper we present a regression-based machine learning approach to email thread summarization. The regression model is able to take advantage of multiple gold-standard annotations for training purposes, in contrast to most work with binary classifiers. We also investigate the usefulness of novel features such as speech acts. This paper also introduces a newly created and publicly available email corpus for summarization research. We show that regression-based classifiers perform better than binary classifiers because they preserve more information about annotator judgements. In our comparison between different regression-based classifiers, we found that Bagging and Gaussian Processes have the highest weighted recall.

• Type: Journal
• Year: 2009
• Where: ICWSM
• Authors: Jan Ulrich, Giuseppe Carenini, Gabriel Murray, Raymond T. Ng
• Keywords: Binary Classifiers, ICWSM 2009, Internet Technology, Multiple Gold-standard Annotations, Regression-based Classifiers