As a guest user you are not logged in or recognized by your IP address. You have
access to the Front Matter, Abstracts, Author Index, Subject Index and the full
text of Open Access publications.
This paper proposes the method of building dictionaries for extracting reputation information. We propose a method of building a highly credible dictionary from the staggering volume of information that customers post on Web pages. The principle of our proposed method is based on assigning a certainty factor which represents the probability that extracted words really do express a feature or evaluation expression and registering only words which have a certainty factor greater than a pre-set threshold value. As a result of experiment, the method of building a highly credible dictionary has been established.