Author Gender Prediction in an Email Stream Using Neural Networks英文文献资料.docVIP

Journal of Intelligent Learning Systems and Applications, 2012, 4, 169-175 1 /10.4236/jilsa.2012.43017 Published Online August 2012 (http://www.SciRP.org/journal/jilsa) Author Gender Prediction in an Email Stream Using Neural Networks William Deitrick, Zachary Miller, Benjamin Valyou, Brian Dickinson, Timothy Munson, Wei Hu * Department of Computer Science, Houghton College, Houghton, USA. * Email: Wei.Hu@ Received March 28 , 2012; revised June 8 , 2012; accepted June 15 , 2012 th th th ABSTRACT With the rapid growth of the Internet in recent years, the ability to analyze and identify its users has become increas- ingly important. Authorship analysis provides a means to glean information about the author of a document originating from the internet or elsewhere, including but not limited to the author’s gender. There are well-known linguistic differ- ences between the writing of men and women, and these differences can be effectively used to predict the gender of a document’s author. Capitalizing on these linguistic nuances, this study uses a set of stylometric features and a set of word count features to facilitate automatic gender discrimination on emails from the popular Enron email dataset. These features are used in conjunction with the Modified Balanced Winnow Neural Network proposed by Carvalho and Cohen, an improvement on the original Balanced Winnow created by Littlestone. Experiments with the Modified Bal- anced Winnow show that it is effectively able to discriminate gender using both stylometric and word count features, with the word count features providing superior results. Keywords: 1-Gram Word Counts; Balanced Winnow; Enron Email; Gender Prediction; Neural Network; Stream Mining, Stylometric Features 1. Introduction 18th century logician Augustus De Morgan was the first to suggest that an aut