Is that a real customer review or just a thinly veiled advert from someone puffing up their own product?
Information engineers in India and Japan believe they have come up with an automatic way to discriminate between personal web pages and commercial pages designed to fool consumers.
The system begins by extracting subjective opinions from a web page, then determines a weighted and categorized ratio of negative to positive expressions to help visitors decide whether or not to trust them.
TechRadar is both good and bad!
It relies on the fact that marketing copywriters and advertisers tend not to report negative comments, whereas genuine personal opinions will be littered with both positive and negative comments.
Personal homepages, blogs, web forum sites and smaller customer opinion sites are regarded as personal pages and generally don't appear high in the search engine results. Finding genuine personal opinions surveys is much harder than finding commercially biased sites, the researchers explain.
The team evaluated the performance of their system using 1200 web pages collected from four categories: products, tourist spots, restaurants, and films. They found that their method is much more effective in finding personal opinion pages than a general search engine, in all categories.
The researchers hope that search engines will adopt their technology. It's a nice idea but hard to see Google jumping at a system that deliberately excludes commercial, ad-paying customers from its search results...
