The modern organization: Translation challenges
Sorah says, "Rosoka needed to be truly multilingual. Many of the existing NLP tools claim to be multilingual, but what they mean is that they have linguistic knowledgebases usually acquired from vendors who provide dictionaries and libraries that make NLP an issue for many licensees. Our current product handles over 200 languages, and we continually strive to improve and expand this capability ...
"In the world of our clients, mixed language documents are important. Documents have to be processed as part of the normal stream, not put in an exception folder and maybe never processed or processed after a delay of hours or days."
My experience with language systems was that most of them produced translations that were "good enough." I can read French, Portuguese and Spanish. The outputs of the systems I have tested were, in my estimation, about 80 to 85 percent accurate.
Other Rosoka innovations that caught my attention were the company's monitoring of certain social media streams to keep track of what languages are most used. In the last sampling period, Sorah told me that English comprises about 40 percent of the Twitter stream.
IMT Rosoka's technology is a blend of computational linguistics and engineering innovation. The blend is a Reese's peanut butter cup approach, according to Sorah. He says, "Greg Roberts, the co-founder of IMT, and I combined competing technologies into a single, unified and unique solution ... my peanut butter (information and statistically-based system and methods) with my partner's chocolate (linguistics-based technologies).
In addition to the translation innovation, Rosoka has a proprietary method that can determine the difference between Tom Ridge, the person and the place, or Paris Hilton, a hotel or the TV star. Rosoka provides language identification, performs entity extraction, generates a language gloss and performs relationship mapping in about one-fourth the time as other translation systems.
Opportunities exist
Despite the presence of Googles and Microsofts in the translation business, there is room for innovation.
Today's business environment is increasingly multilanguage. Human translation is expensive. In Harrod's Creek, Kentucky, where I live, the automatic teller machine offers a language choice. The linguistic monoculture in rural Kentucky has changed and quickly. Organizations operating in larger urban areas have to adapt to a multilanguage world. Despite the troubled economic climate, there are opportunities for giants like Google and innovators like IMT.