Author_Institution :
Dept. of Training & Teaching, Army Officer Acad. of PLA, Hefei, China
Abstract :
Operation-oriented Document Natural Language Understanding (take ODNLU for short) is an important approach to automatic plotting research. However, current researches have not given a feasible method to ODNLU, but with some designed processes. The purpose of this paper is to achieve ODNLU on the event level. According to the need of automatic plotting, the event model is proposed, which contains four different classic events: configuration event, constitution event, task event, and coreference event. It describes the composition of document, and the relationship between military subjects. Then, the model identification method based on BayesNet algorithm is presented. On the basis of these analyses, the whole ODNLU process is designed, consisting of word segment, semantic role labeling, and event model analysis. The experimental results show that this ODNLU method is feasible and effective, which achieves an average precision at 89. 9%
Keywords :
Bayes methods; document handling; natural language processing; BayesNet algorithm; ODNLU; configuration event; constitution event; coreference event; event model analysis; model identification method; operation-oriented document natural language understanding; semantic role labeling; task event; word segment; Algorithm design and analysis; Analytical models; Feature extraction; Labeling; Natural languages; Semantics; Training; BayesNet algorithm; automatic plotting; event model; natural language understanding; semantic role labeling;
