We propose a system that can score online reputation text for a given attribute, e.g., design and portability, in a certain category, e.g., mp3 players and laptops. Additionally, we visualize scored results as an interactive system. The scoring system has evaluative expression dictionaries created in advance using reviews. They comprise evaluative values that determine the attribute-wise polarity of evaluative expressions: good or bad. Our system analyzes online reputation using the evaluative expression dictionaries, and then outputs the scores of the reputation. The system helps users to ascertain a reputation easier than merely by reading through textual information. ?? 2014 Springer International Publishing Switzerland.
CITATION STYLE
Tanimoto, Y., Niitsuma, H., & Ohta, M. (2014). Prototype system for visualizing review text using evaluative expression dictionaries. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8709 LNCS, pp. 662–665). Springer Verlag. https://doi.org/10.1007/978-3-319-11116-2_65
Mendeley helps you to discover research relevant for your work.