IBM will have taken Computer fluent in the human
Artificial intelligence-systems to process an increasing role in business-how your skills and learn from large volumes of data have improved. But they have struggled to understand the types of nuances that appear in everyday communication between people.
IBM ( is the process of adding new tools to your AI system Watson, which can fix the problem, and to better understand the human language. The “natural language processing” technology was developed for the IBM Research project, Debater, the AI machine, the first to compete against a world-class human debater in the last year. )
in a debate, project-Debater needs to have the ability to , to hear, to understand his opponents, his or her reasoning, and then formulate a response quickly, all without internet access. this meant That the computer needs to detect the location and the sense of Proverbs and idioms — as well as certain dialects or industry-specific terms like “open a can of worms” or “a little helpful” (earlier, an AI system could think, someone was actually opening a can of worms).
the use of companies now in the location of IBM-natural language processing tools for more efficient and effective way to comb through documents and research to improve the field of in-depth communication and customer service.
“Everything that happens in a business is based on communication between employees, business partners, customers, and the communication is in the form of E-Mails, chats, texts, documents,” Rob Thomas, general manager of IBM Data and AI, told CNN Business. “Some of the best data that a company has, is codified in all the language, all of the communications.”
Around 48% of the global chief investment Officer search, to provide AI systems in your company this year, after a year in 2019, a survey by the market research Institute Gartner. January IBM survey of 4,500-technology “decision makers” in companies around the world in a similar way, found that 45% of companies have adopted, with more than 1,000 employees, AI.
A great improvement over existing AI systems is the ability of the system to “sentiment analysis” — to see what someone or written has to say and understand what you really are trying to communicate, and what is the context. For example, could understand it, but now that someone says:,, you have “cold feet” that you are concerned about, not that her limbs are cool.
“A lot of how we speak and interact, certainly in customer service, is twists and turns with a lot of speech, or of the terms of a certain industry,” said Thomas.
When this ability is on an existing IBM AI-tool called Watson Discovery, it is also due to make digging easier for companies, and can collect evidence from finds, documents, and other written communications.
The system is better able to understand, to classify the Central themes or important points in documents and give them to more specific categories, so that you are more useful. It can also generate concise summaries of mass amounts of data. The tool can also tell if two documents make essentially the same point with different words.
“If Watson goes over a million documents that it can recognize: What are the information, the most relevant is given, the problem is trying to solve?” Thomas said.
This could be useful, for example, if a company fielding customer service requests and needs to you, you will be treated to different places, so that is correct. It could also be used by a law firm to sort through the documents and come to conclusions about you for the discovery in a legal case — what is normally a lawyer, for days or weeks, the machine would be a matter of minutes.
One of the main concerns about the field of artificial intelligence is that, in some cases, such systems have been found to show biases towards certain groups of people, including people of color.
Thomas stressed that trust is essential for a company, the KI, as well as their customers. He said, IBM has a tool called Watson open scale that, in addition to other functions, you can do that bias the detection and prevention of company ” AI-tools.
“trust is important — be able to use a KI, you can explain that is free of bias, that is trustworthy,” said Thomas.
And while the artificial intelligence of the system, kind can change how people work, they are unlikely to negate the need for human workers. Instead, said Thomas, the tools, people could make more effective at their jobs.
“Maybe the problem is too hard for the AI, but if you have a few of these tech with (a human worker), your ability and speed to find the solution quickly,” he said.