Abstract:
With the increased amount of information
rapidly available on the World Wide Web,
Internet users that want to know opinions about
products are becoming difficult to determine
which product (entity) is the best on many
product sites. When the product manufacturers
are interesting how the product compares with
those of competitors, opinion mining on
comparative sentences becomes very important.
Mining on comparative sentences is called
comparison mining. The purpose of this paper is
to get the best entity from superlative relations in
the comparison mining. This paper focuses on
mining comparative (opinion) words and
determines the best entity on comparative
sentences from the product reviews data set.
Determining the best entity depends on just one
feature that has same nature or application
domain. This paper mentions a rule- based
approach that integrates two sequential rule
mining techniques that utilizes POS tagging.
Determining the best entity on comparative
sentences is effective and time saving, not only
for individuals but also for organizations such as
business intelligence units.
