A new approach to ontology-based semantic modelling for opinion mining

Alfrjani, R., Osman, T. ORCID: 0000-0001-8781-2658 and Cosma, G. ORCID: 0000-0002-4663-6907, 2016. A new approach to ontology-based semantic modelling for opinion mining. In: D. Al-Dabass, A. Orsoni, R. Cant ORCID: 0000-0001-9610-7205 and G. Jenkins, eds., UKSim2016: Proceedings of the UKSim-AMSS 18th International Conference on Mathematical Modelling & Computer Simulation, Cambridge, 6-8 April 2016. Institute of Electrical and Electronics Engineers (IEEE), pp. 267-272. ISBN 9781509008872

[img]
Preview
Text
PubSub5149_Cosma.pdf - Published version

Download (394kB) | Preview

Abstract

With the fast growth of World Wide Web 2.0, a great number of opinions about a variety of products have been published in blogs, forums, and social networks. Opinion mining tools are needed to enable users to efficiently process a large number of reviews found online, in order to determine the underlying opinions. This paper presents a new methodology for semantic modelling of the domain knowledge for opinion mining. In particular, the new methodology focuses on modelling the domain knowledge in such a way that it can be translated to a formal ontology, which can then be automatically enriched with ground facts obtained from public Linked Open Data resources. The methodology also considers procedures to link between the formal ontology and Natural Language Processing. Our approach successfully enriches the ontology with the relevant ground facts. This ontology can then be used to perform a variety of data mining tasks including sentiment analysis and information retrieval.

Item Type: Chapter in book
Creators: Alfrjani, R., Osman, T. and Cosma, G.
Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Date: 2016
Divisions: Schools > School of Science and Technology
Depositing User: Linda Sullivan
Date Added: 10 May 2016 13:20
Last Modified: 09 Jun 2017 14:25
URI: http://irep.ntu.ac.uk/id/eprint/27758

Actions (login required)

Edit View Edit View

Views

Views per month over past year

Downloads

Downloads per month over past year