OPINION MINING BASED ON MODAL MEANINGS
Published: 25 Aug 2014
Abstract: This research aims to classify Korean modal expressions influencing a sentiment value. With frequent and easy access to the web, user-generated online texts are easily created in great quantities and usually contain opinions which express the user’s sentiments. Sentiments are identified from a given opinion and evaluated to determine the final polarity value of an entity occurring in the text. However, the accuracy of evaluating the polarity is not satisfying if the number of sentiment words appearing in the text is simply counted to interpret their semantic orientations. In particular, polarity values are affected by contextual words or expressions, some of which are represented utilizing modality. The modal expressions are constructed with mood or a modal system in Korean. The irrealis in particular convert a sentiment value to a hypothetical or unlikely state, in which interpretation of the sentiment has to be approached differently. Modality affects the polarity value of a sentiment in various ways which include diminishing, intensifying or removing the polarity value at all. Therefore, correctly interpreting modality plays a crucial role in assigning an accurate polarity value. This paper focuses on classifying Korean modal expressions which influence interpreting sentiments. Their structures are also discussed since Korean modality presents a complex predicate, utilizing modal suffixes, auxiliaries, and modal verbs.
Keywords: sentiment analysis, polarity value, modal meaning, mood, modal systems
Download full text
Back to the contents of the volume
© 2018 The Author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License http://creativecommons.org/licenses/by/3.0/
, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. This permission does not cover any third party copyrighted material which may appear in the work requested.