MALLET

From the MALLET homepage : MALLET is a Java-based package for statistical natural language processing, document classification, clustering, topic modeling, information extraction, and other machine learning applications to text.

We chose to represent two distinct topics within the letters we obtained in Special Collections at Washington and Lee University using topic modeling. Used for this purpose, MALLET searches for topics in data set.
___________________________________

TOPIC 1

lee university future women study hope raised position member successful idea respond expect fewer individuals top obtaining greatly conviction

Analysis

This model focuses on the study concerning Washington and Lee’s  future contending with other top universities in the United States. With the falling rate of college-going males and academic quality at W&L, John D. Wilson and the Board of Trustees aimed to respond in a timely manner to address these concerns to secure a successful future for the university. Their solution to this concern was coeducation.

TOPIC 2

alumni board president quality male experience student change people education support opinion found encourage size important age place present

Analysis

This model concerns how the educational environment and experience would change if women were to come to Washington and Lee University.  Continuing academic excellence of W&L’s past involved questions of the school’s surely changing academic and atmospheric qualities in 1985 with the switch to coeducation.

___________________________________
Source: McCallum, Andrew Kachites. “MALLET: A Machine Learning for Language Toolkit.” http://mallet.cs.umass.edu. 2002.