IBM plans to integrate its AI Project Debater into Watson products.

IBM Research’s Project Debater, the only AI system capable of debating humans on complex topics, will soon help businesses to better utilize language and text-based data. The Debater technology enabled a computer to read up documents and articles and enter a debate with a human being. IBM (NYSE: IBM) will soon commercialize this AI-powered technology through its products. These include Watson NLP, Watson Knowledge Studio, and Watson Discovery. (CNET)

Businesses and individuals can use this technology in practical applications to identify trends in large volumes of data. Typical examples of such data are tweets, internet blogs, and podcasts.

“Language is a tool for expressing thought and opinion, as much as it is a tool for information,” said Rob Thomas, General Manager, IBM Data and AI in a statement. “This is why we’re harvesting technology from Project Debater and integrating it into Watson – to enable businesses to capture, analyze, and understand more from human language and start to transform how they utilize intellectual capital that’s codified in data.”

AI bridging the gap between humans and computers

AI tools such as neural networks and deep learning enable computers to move beyond ordered data (such as spreadsheets) into an unstructured human world. Machines are therefore able to make sense of it. These are applications such as self-driving cars and scanning millions of images. It could also be understanding a human being at the other end of a telephone call.

The human language, with its antonyms, synonyms, idioms, and metaphors has always been a challenge for computers.

For instance, a computer can hardly be blamed for misunderstanding the idiom “a can of worms.”

It would normally assume that the text referred to a physical tin containing worms.

Three specific applications

The inclusion of Debater’s key Natural Language Processing (NLP) capabilities in Watson means that users can perform operations such as sentiment analysis and summarizing information. For example, the system can judge incoming data to understand whether people are reacting positively or negatively to an event, e.g. in tweets and social media posts.

In an example of a summarization application, IBM tested the technology at this year’s Grammy music awards. The system gleaned capsule biographies and summaries of fan views of artists. The input data comprised 18 million blog posts, articles, and biographies.

In another example, ESPN Fantasy Football analyzes millions of football data sources each day during the season. It can, therefore, offer real-time insights to millions of fantasy football players. For example, the “Player Buzz” section summarizes the positive or negative comments about a player.

The system can also help to classify language-based data into “clusters.” For example, at French bank Crédit Mutuel, the system was able to route incoming calls to its contact center to specific reps instead of to a “general inquiry” queue.

Related Story:   Artificial Intelligence: How Facebook Uses AI to Detect Fake Accounts