Wykorzystanie metody opartej na wzorcach w automatycznej analizie opinii konsumenckich
Using pattern-based opinion mining
Author(s): Janusz Tuchowski, Katarzyna WójcikSubject(s): Economy
Published by: Wydawnictwo Uniwersytetu Ekonomicznego we Wrocławiu
Keywords: text-mining; Web-mining; taxonomy; classification of text documents; opinion mining; sentiment analysis; patterns; Spejd
Summary/Abstract: Sentiment analysis or opinion mining is a field of research that can have a significant impact on today’s business. The increasing number of consumers’ reviews created the need of its automatic analysis. This issue is gaining popularity for both – researchers and entrepreneurs, for whom consumers’ reviews are an important source of business information. The main aim of this paper is to examine pattern-based classification of opinions. Pattern-based approach allows identifying certain phrases in opinions to which sentiments can be assigned. An advantage of this approach is a possibility to detect phrases that modify sentiment like negation, nullification, strengthening and others. The approach used in the research is compared with other approaches to opinions classification.
Journal: Prace Naukowe Uniwersytetu Ekonomicznego we Wrocławiu
- Issue Year: 2015
- Issue No: 385
- Page Range: 314-324
- Page Count: 11