Please use this identifier to cite or link to this item: http://repository.kln.ac.lk/handle/123456789/20163
Title: Part of speech tagging for Twitter conversations using Conditional Random Fields model
Authors: Fanoon, A.R.F.S.
Uwanthika, G.A.I.
Keywords: Bigram model
Conditional Random Field
Tagging
Unigram model
Issue Date: 2019
Publisher: IEEE International Research Conference on Smart computing & Systems Engineering (SCSE) 2019, Department of Industrial Management, Faculty of Science, University of Kelaniya, Sri Lanka
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
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
URI: http://repository.kln.ac.lk/handle/123456789/20163
Appears in Collections:Smart computing & Systems Engineering - (SCSE - 2019)

Files in This Item:
File Description SizeFormat 
SC-1 (17).pdf2.01 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.