Tibetan Base Noun Phrase Identification Framework Based on Chinese-Tibetan Sentence Aligned Corpus

Proceedings of COLING 2012: Technical Papers, pages 2141–2158, COLING 2012, Mumbai, December 2012. Tibetan Base Noun Phrase Identification Framework Based on Chinese-Tibetan Sentence Aligned Corpus NUO Ming Hua1,2 LIU Hui Dan1,2 ZHAO Wei Na1,3 MA Long Long1 WU Jian1 DING Zhi Ming1 (1) Institute of Software, Chinese Academy of Sciences, Beijing, China (2) Graduate University of the Chinese Academy of Sciences, Beijing, China (3) Qinghai Normal University, Xining, China Minghua,huidan,weina,longlong,wujian,zhiming@iscas.ac.cn ABSTRACT This paper presents an identification framework for extracting Tibetan base noun phrase (NP). The framework includes two phases. In the first phase, Chinese base NPs are extracted from all Chinese sentences in the sentence aligned Chinese-Tibetan corpus using Stanford Chinese parser. In the second phase, the Tibetan translations of those Chinese NPs are identified using four different methods, that is, word alignment, iterative re-evaluation, dictionary and word alignment, and sequence intersection method. We implemented and tested these methods on Chinese-Tibetan sentence aligned unlabelled corpus without Tibetan POS tagger and Treebank. The experimental results demonstrate these methods can get satisfactory results, and the best performance with 0.5283 precision is got using sequence intersection identification method. The identification framework can also be extended to extract Tibetan verb phrase. Title and abstract in Chinese 基于汉藏句子对齐语料的藏文BaseNP识别框架 本文提出藏文BaseNP识别框架，它分两步完成。先通过句法分析得到汉语BaseNP。再为 这些汉语BaseNP从汉藏句子对齐语料中识别出藏文对应短语。我们应用四种方法识别藏 文BaseNP，分别是词对齐、迭代重估算法、词典和词对齐相结合的方法以及基于序列相 交的方法。评价实验表明，没有藏文词性标注和树库的前提下，基于序列相交的方法性能 最好。本文提出的框架可以用于藏文动词短语识别任务中。 KEYWORDS : Tibetan information processing; base noun phrase; head-phrase; CHINESE KEYWORDS :藏文信息处理,基本名词短语,中心语块 2141 1 Introduction Shallow parsing identifies the non-recursive cores of various phrase types in text, possibly as a precursor to full parsing or information extraction (Abney, 1991). The paradigmatic shallow parsing problem is NP chunking, which finds the non-recursive cores of noun phrases called BaseNPs. It can help to solve many natural language processing tasks, such as information extraction, named entity extraction, machine translation, and text summarization and so on. In general, researchers consider chunking as a kind of tagging problem or a sequence labelling task. Machine learning techniques are often applied to chunking. In Tibetan information processing, the shortage of Tibetan language resource leads to the fact that most of the techniques related text processing are still developing. Since 2003, research on Tibetan corpus, Tibetan word segmentation are reported. Although there is no public available Tibetan annotated corpus and Tibetan Treebank, we intend to extract Tibetan BaseNP using machine translation techniques based on our Chinese-Tibetan sentence aligned corpus. The research on Tibetan BaseNP is still in the initial stage. So far, there is no related report. In this paper, we will propose several methods for automatic identification of Tibetan base noun phrases. The concept of BaseNP is initially put forward by (Church, 1998). In English, BaseNP is simple and non-recursive noun phrase which does not contain other noun phrase descendants. It cannot meet the needs in Tibetan information processing. Presently, different definitions of Chinese BaseNP are used on the basis of research field. According to our Chinese-Tibetan corpus, restrictive attribute phrase is in the scope of our BaseNP extraction. Observing that the Tibetan BaseNP is different from English, BaseNP in Tibetan can be recursively defined as follows, which is …

Content truncated to 500 words