A Hybrid Approach to the Sentiment Analysis Problem at the Sentence Level

Orestes Appel, Francisco Chiclana, Jenny Carter, Hamido Fujita

Research output: Contribution to journalArticle

91 Citations (Scopus)


This contribution presents a hybrid approach to Sentiment Analysis (SA) encompassing the use of semantic rules, fuzzy sets, unsupervised machine learning techniques and a sentiment lexicon improved with the support of Senti-WordNet. A Hybrid Standard Classification is first carried out, which is further enhanced into a Hybrid Advanced approach incorporating linguistic classification of semantic polarity modelled using fuzzy sets. The mechanism of the new SA methodology is illustrated by applying it to compute the polarity of a given sentence and to a benchmarking publicly available dataset: the Movie Review Dataset.
Original languageEnglish
Pages (from-to)110-124
Number of pages15
JournalKnowledge-Based Systems
Early online date20 May 2016
Publication statusPublished - 15 Sep 2016
Externally publishedYes


Cite this