Design of a POS tagger using conditional random fields for Malayalam

Author

Krishnapriya, V. ; Sreesha, P. ; Harithalakshmi, T.R. ; Archana, T.C. ; Vettath, Jayasree N.

Author_Institution

Dept. of Comput. Sci. & Eng., Sreepathy Inst. of Manage. & Technol., Palakkad, India

fYear

2014

fDate

17-18 Dec. 2014

Firstpage

370

Lastpage

373

Abstract

Parts of Speech tagging, is a process of marking the words in a text as corresponding to a particular part of speech, based on its definition and context POS tagger plays an important role in Natural language applications like speech recognition, natural language parsing, information retrieval and extraction. This paper discusses architecture for designing a Part-Of-Speech (POS tagger for Malayalam language using Conditional Random Field (CRF). The experiments presented in this paper use an annotated corpus of 1028 sentences (11,315 words) and tagset consists of 100 tags. A trigram based tagging scheme is involved in the experiments. The proposed system is based on an empirical approach that models the human POS tagging processing more realistically than the existing systems, without compromising the efficiency and accuracy.

Keywords

information retrieval; natural language processing; speech processing; CRF; Malayalam language; POS tagger design; annotated corpus; conditional random field; conditional random fields; information extraction; information retrieval; natural language applications; natural language parsing; speech recognition; speech tagging; text words; Accuracy; Hidden Markov models; Speech; Speech processing; Support vector machines; Tagging; Training; CRF; Hidden Markov Model; Malayalam; POS tagging; Stochastic process;

fLanguage

English

Publisher

ieee

Conference_Titel

Computational Systems and Communications (ICCSC), 2014 First International Conference on

Conference_Location

Trivandrum

Print_ISBN

978-1-4799-6012-5

Type

conf

DOI

10.1109/COMPSC.2014.7032680

Filename

7032680