Natural Language Understanding Becoming a Bigger Part of Our Lives
May 01, 2012
By Steve Anderson, Contributing TMCnet Writer
There's no denying that Siri is changing the way people look at their hardware. But as much a game-changer as Siri is by itself, it's the basis behind Siri—natural language understanding—that's really shaking the game up as we know it.
Natural language understanding systems have been something of a hot commodity since well before Siri showed up in the market. Those who remember the Watson supercomputer, which managed to make the two greatest Jeopardy champions in history look like impressively detailed statuary by comparison when it took them on in a recent match, have already seen the impact of systems that can work with a natural language understanding system.
Basically, people like the thought of a computer system that allows them to ask questions, or issue commands, in normal everyday language and have meaningful responses returned to them. The old days of computer use, in which specific commands were required to execute programs and receive the desired output, are rapidly receding in favor of systems that can work to a normal person's style of use.
However, not everyone shares that assessment. Many have questioned the value of natural language understanding in the consumer market, at least in the stages it's currently in, calling it too expensive to properly design and implement. To some, a proper implementation of natural language understanding systems has a long way to go. But considering the number of products that are currently on the market and making money--products like Siri, or Nuance Communcations' lineup of software products—the assertions of those speaking against natural language understanding could be seen as carrying less weight.
But the actual state of natural language understanding may be found in the middle. Indeed, natural language understanding isn't always the best. Using the Siri example again, many have found that there are several things Siri can't, or won't, do. But that's not to say that future versions of Siri won't pick up the slack of its current generation, and people are certainly finding that the current version of Siri already does plenty in its current state.
With the current standard of natural language understanding software already doing some impressive things in the market space, and the future ahead looking bright by any measure, natural language understanding may not be all it could be yet, but the idea that it could be more—and not in the far future, but much nearer-term than some expect—is certainly not out of line. Thus, natural language understanding systems are certainly worth investigation and development.
Edited by Juliana Kenny