Treebank based deep grammar acquisition and Part-Of-Speech Tagging for Sanskrit sentences
Treebank based deep grammar acquisition and Part-Of-Speech Tagging for Sanskrit sentences
复制标题
基于树库的梵文句子深度语法获取和词性标注
DOI：
10.1109/conseg.2012.6349476
复制
发表时间：
2012
期刊：
影响因子：
--
通讯作者：
Suresh Jain
中科院分区：
文献类型：
--
作者：
N. Tapaswi;Suresh Jain
几千年来，梵语一直是印度的东方语言。它是大多数印度语言的基础。自然语言句子中固有的歧义性。在这里，一个词可以有多种含义。 Morphology process takes word in isolation and fails to disambiguate correct sense of a word. Part-Of-Speech Tagging (POST) takes word sequences in to consideration to resolve the correct sense of a word present in the given sentence.高效的 POST 已开发用于处理英语、日语和中文，但缺乏印度语言。在本文中，我们的工作为梵语语言提供了简单的基于规则的 POST。 It uses rule based approach to tag each word of the sentence.这些规则存储在数据库中。 It parses the given Sanskrit sentence and assigns suitable tag to each word automatically.我们已经针对该语言的 15 个标签和 100 个单词测试了这种方法，该基于规则的标记器为给定句子中的所有变形单词提供了正确的标签。
Sanskrit since many thousands of years has been the oriental language of India. It is the base for most of the Indian Languages. Ambiguity is inherent in the Natural Language sentences. Here, one word can be used in multiple senses. Morphology process takes word in isolation and fails to disambiguate correct sense of a word. Part-Of-Speech Tagging (POST) takes word sequences in to consideration to resolve the correct sense of a word present in the given sentence. Efficient POST have been developed for processing of English, Japanese, and Chinese languages but it is lacking for Indian languages. In this paper our work present simple rule-based POST for Sanskrit language. It uses rule based approach to tag each word of the sentence. These rules are stored in the database. It parses the given Sanskrit sentence and assigns suitable tag to each word automatically. We have tested this approach for 15 tags and 100 words of the language this rule based tagger gives correct tags for all the inflected words in the given sentence.