Dabhurbayar & Bayarmend. (2010). Recognizing Basic Verb Phrases in Mongolian Corpus. Altai Hakpo 20: 55-70. The Altaic Society of Korea.

Grammatically, phrase analysis is a key process for completing the syntactic analysis. It is also an important topic for Mongolian language information processing. Phrasal information is very useful for applications of Mongolian language technology. For example, Mongolian machine translation, Mongolian input system, Mongolian text mining. This paper described some features of Mongolian basic verb phrases (BVP) including morphological, POS, subcategorization and contextual information. These features were integrated into the syntactic structural rules and then a rule‐based tagger was developed to delimit and distinguish BVPs in Mongolian corpus. The result of test indicated in table 7 shows that this rule‐based phrase tagger can tag correctly most of the BVPs in our Mongolian corpus. In this test, there appeared a lot of structural ambiguity. It is important to import semantic and contextual knowledge systematically for the disambiguation. We think that it is also necessary to consider the relations between all kinds of phrases for developing a robust phrase tagger.

Keywords: Mongolian language, corpus, basic verb phrases, analysis