Authoritative sources in a hyperlinked environment
Journal of the ACM (JACM)
Mining the peanut gallery: opinion extraction and semantic classification of product reviews
WWW '03 Proceedings of the 12th international conference on World Wide Web
Opinion observer: analyzing and comparing opinions on the Web
WWW '05 Proceedings of the 14th international conference on World Wide Web
EC '06 Proceedings of the 7th ACM conference on Electronic commerce
Modeling and analyzing review information on the web focusing on credibility
Proceedings of the 2009 ACM symposium on Applied Computing
Searching coordinate terms with their context from the web
WISE'06 Proceedings of the 7th international conference on Web Information Systems
Application of semi-supervised learning to evaluative expression classification
CICLing'06 Proceedings of the 7th international conference on Computational Linguistics and Intelligent Text Processing
Finding Comparative Facts and Aspects for Judging the Credibility of Uncertain Facts
WISE '09 Proceedings of the 10th International Conference on Web Information Systems Engineering
Evaluating truthfulness of modifiers attached to web entity names
WAIM'10 Proceedings of the 11th international conference on Web-age information management
Hi-index | 0.00 |
The value of a brand name is an important factor that consumers often take into consideration when making their purchasing decisions. However, it is difficult for users to evaluate correctly the value of a brand name, especially when they encounter it for the first time. In reality, sometimes a brand's description or its use is purposely manipulated so as to give an impression of high value. In another way, a non-existing brand name may be used to attract consumers. We call such names "glorified terms." In this paper, we propose a method for evaluating a brand's value from texts on the Web. To this end, we first acquire candidates of attributes useful for evaluating whether a term is a brand name or a glorified term. The candidates are evaluated according to the idea whereby explanations about a real brand name often contain attributes describing its quality. We implemented a prototype system especially for agricultural and livestock products. The system judges whether a given one is a glorified term or a well-known brand name from several viewpoints. We conducted preliminary experiments and we achieved 74% - 85% accuracy rate.