No one knows exactly how many blogs are being posted on the Internet, but it’s a number that grows daily, making blogging an increasingly important venue for the public discussion of everything-including products. Software now in beta use among a limited set of testers and scheduled for commercial launch in October can pull language from web logs-narratives usually maintained by individuals that recount their experiences and their reactions to those experiences-and offer manufacturers, retailers and other businesses a window on what consumers are saying about specific products and services, Matt Rice, founder of software maker Blabble, tells Internet Retailer.
Rice notes that some search engines such as Google pull listings from blogs, but they treat the blog listing as they would any web page, requiring the searcher who specifically is seeking blog content to read through all the listings returned in a search to identify those from blogs. Existing software products aggregate listings from blogs, but require the user seeking a view of overall trends or opinions as represented in blogs to read through all the blog listings to make that determination manually.
Rice says Blabble goes a step farther by incorporating natural language processing that parses blog listings returned in a search into parts of speech so as to extract from them words, phrases and constructions that indicate opinion. “50,000 people may write about a topic, but you don’t have time to read 50,000 listings,” says Rice. “And I probably don’t care about one individual opinion; it’s the aggregate that I care about."
Blabble doesn’t crawl the web in response to a search, instead drawing from its own searchable database that aggregates content from what are currently more than 2 million blog authors. To fill and grow its database, it secures lists of blogs and also draws from companies that provide consumers with software and platforms for creating blogs. Blabble gets a direct feed from such companies so every change or addition to a posted blog is captured, which keeps its database current.
The software can search the database for relevant words, phrases, and even within defined time frames to gauge how often bloggers referred to a product by hour, day, week and more. It also can search for the same idea as expressed by phrases composed of different language. It encompasses a “tone engine” that, for example, ranks adjectives and descriptions on a scale to determine the degree of a blogger’s negative or positive feeling about the product or service being discussed. “For example, the statement ‘that was the greatest movie I ever saw’ might get a ranking of nine, while ‘that was an okay movie’ might get a two,” Rice says.
Rice says the natural language processing underlying the software can save those mining blogs for trends or opinions “hundreds of hours” of research, the concept being that it brings more automation to the qualitative search of blog content, a process that currently is largely manual. Blabble is being offered free of charge while in the beta stage; users will pay a monthly fee starting in October when the product officially launches, Rice adds.