1. UKM Learning and Research Repository
2. Master Thesis / Tesis Master
3. Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

Title:	Hybridised N-gram model using naive bayes for classification of political sentiments on twitter
Authors:	Jamilu Awwalu (P72257)
Supervisor:	Azuraliza Abu Bakar, Prof. Dr.
Keywords:	Computational linguistics -- Online social networks -- Political aspects Information theory -- Online social networks -- Political aspects Universiti Kebangsaan Malaysia -- Dissertations Dissertations, Academic -- Malaysia
Issue Date:	12-Jul-2015
Description:	Twitter, an online micro-blogging and social networking service provides registered users the ability to write in 140 characters anything they wish; hence providing them the opportunity to express their opinions and sentiments on events taking place. Politically sentimental tweets are top trending tweets whenever election is by the corner, users tweet about their favourite candidates or political parties and at times give their reasons for that. In this study, Katz Back-off is used for hybridization of N-gram models after applying Laplace smoothing to Naive Bayesian classifier. This was done as a means of smoothing and addressing the limitation of accuracy in terms of precision and recall of N-gram models caused by the 'zero count problem' in N-gram models. Result from the baseline model shows an increase of 6.05% in average F-Harmonic accuracy in comparison to the N-gram model, and 1.75% increase in comparison to the Semantic-Topic model proposed from a previous study on the same dataset i.e. Obama-McCain dataset.,Master of Information Technology
Pages:	79
Call Number:	P98.A976 2015 3 tesis
Publisher:	UKM, Bangi
Appears in Collections:	Faculty of Information Science and Technology / Fakulti Teknologi dan Sains Maklumat

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