Part of speech tagging for Twitter conversations using Conditional Random Fields model

No Thumbnail Available

Date

2019

Journal Title

Journal ISSN

Volume Title

Publisher

IEEE International Research Conference on Smart computing & Systems Engineering (SCSE) 2019, Department of Industrial Management, Faculty of Science, University of Kelaniya, Sri Lanka

Abstract

Part-of-Speech Tagging is the technology of assigning the appropriate parts-of-speech to a word. Part-of-speech tagging is very useful in information retrieval, information extraction, and speech processing. This research presents a part-of-speech tagging, especially for twitter text data. The process of part-of-speech tagging for twitter conversation is a difficult task. Several approaches have been made to develop an accurate tagging system but most of them are relevant to news text data and web contents. Therefore, this research intends to develop a part-of-speech tagger model for twitter speech. using CRF toolkit. The system was developed for nearly 1000 twitter conversations employing Conditional Random Field stochastic model. The data for twitter speech was downloaded from the internet. A POS-tagged text corpus, template file and CoNLL file for both training and testing database were prepared accordingly. The training was carried out for both the unigram model as well as the bigram model. The performance of the system over these models was obtained through this examination which showed a significant efficiency, calculated from the number of correctly tagged words and the total number of words

Description

Keywords

Bigram model, Conditional Random Field, Tagging, Unigram model

Citation

Fanoon, A.R.F.S. and Uwanthika, G.A.I. (2019). Part of speech tagging for Twitter conversations using Conditional Random Fields model. IEEE International Research Conference on Smart computing & Systems Engineering (SCSE) 2019, Department of Industrial Management, Faculty of Science, University of Kelaniya, Sri Lanka.P.108

Endorsement

Review

Supplemented By

Referenced By