TRDDC @ FIRE 2013: System for Classification of Legal Propositions Report for Legal Track at FIRE 2013
暂无分享,去创建一个
In this work, we describe a system for classification of legal propositions using a multi-class maximum entropy classifier. We designed various features to capture the underlying characteristics of legal propositions. The best performing feature set was chosen by 10-fold cross-validation over the training set. The system achieved the best F-measure of 62.7%.
[1] Adam L. Berger,et al. A Maximum Entropy Approach to Natural Language Processing , 1996, CL.
[2] Claire Grover,et al. Sentence classification experiments for legal text summarisation , 2004 .