A Unified Statistical Model for the Identification of English BaseNP

A Unified Statistical Model for the Identification of English BaseNP Endong Xun Microsoft Research China No. 49 Zhichun Road Haidian District 100080, China, i-edxun@microsoft.com Ming Zhou Microsoft Research China No. 49 Zhichun Road Haidian District 100080, China, Mingzhou@microsoft.com Changning Huang Microsoft Research China No. 49 Zhichun Road Haidian District 100080, China, cnhuang@microsoft.com Abstract This paper presents a novel statistical model for automatic identification of English baseNP. It uses two steps: the N- best Part-Of-Speech (POS) tagging and baseNP identification given the N-best POS-sequences. Unlike the other approaches where the two steps are separated, we integrate them into a unified statistical framework. Our model also integrates lexical information. Finally, Viterbi algorithm is applied to make global search in the entire sentence, allowing us to obtain linear complexity for the entire process. Compared with other methods using the same testing set, our approach achieves 92.3% in precision and 93.2% in recall. The result is comparable with or better than the previously reported results. 1 Introduction Finding simple and non-recursive base Noun Phrase (baseNP) is an important subtask for many natural language processing applications, such as partial parsing, information retrieval and machine translation. A baseNP is a simple noun phrase that does not contain other noun phrase recursively, for example, the elements within [...] in the following example are baseNPs, where NNS, IN VBG etc are part-of-speech tags [as defined in M. Marcus 1993]. [Measures/NNS] of/IN [manufacturing/VBG activity/NN] fell/VBD more/RBR than/IN [the/DT overall/JJ measures/NNS] ./. Figure 1: An example sentence with baseNP brackets A number of researchers have dealt with the problem of baseNP identification (Church 1988; Bourigault 1992; Voutilainen 1993; Justeson & Katz 1995). Recently some researchers have made experiments with the same test corpus extracted from the 20th section of the Penn Treebank Wall Street Journal (Penn Treebank). Ramshaw & Markus (1998) applied transform- based error-driven algorithm (Brill 1995) to learn a set of transformation rules, and using those rules to locally updates the bracket positions. Argamon, Dagan & Krymolowski (1998) introduced a memory-based sequences learning method, the training examples are stored and generalization is performed at application time by comparing subsequence of the new text to positive and negative evidence. Cardie & Pierce (1998 1999) devised error driven pruning approach trained on Penn Treebank. It extracts baseNP rules from the training corpus and prune some bad baseNP by incremental training, and then apply the pruned rules to identify baseNP through maximum length matching (or dynamic program algorithm). Most of the prior work treats POS tagging and baseNP identification as two separate procedures. However, uncertainty is involved in both steps. Using the result of the first step as if they are certain will lead to more errors in the second step. A better approach is to consider the two steps together such that the final output takes the uncertainty in both steps together. The approaches proposed by Ramshaw & Markus and Cardie&Pierce are deterministic and local, while Argamon, Dagan & Krymolowski consider the problem globally and assigned a score to each …

Content truncated to 500 words