IBM Unveils New NLP Features for Watson Discovery

IBM said it is adding new natural language processing features to Watson Discovery, its AI search and text analytics platform.

The new features unveiled on Wednesday allow customers to train Watson Discovery to read, understand and display accurate answers from large, complex and industry-specific skills without the need for computer science skills.

Watson Discovery’s new Smart Document Understanding feature includes a pre-trained model that automatically understands the visual structure and layout of a document, with no additional input from a developer or data researcher. The feature is available now in the Plus, Enterprise and Premium plans.

“Out of the three new planned features that will be infused into IBM Watson Discovery … I think users will be most attracted to intelligent document understanding,” said Ritu Jyoti, an analyst at IDC. “Time to value is crucial in an age of AI adoption and lack of appropriate skills and talent.”

“This capability helps achieve agility and reduction in additional costs and resources,” she added.

Another new feature is Automatic Text Pattern Recognition, an advanced pattern creation tool currently available in beta for Plus, Premium and Enterprise customers.

The feature allows the user to quickly identify business-specific text patterns in documents.

Watson Discovery also includes a new custom device extraction feature, now in beta, that allows companies to train NLP (natural language processing) models to identify company-specific words and phrases.

Jyoti said this feature reduces the effort required for data preparation, simplifies labeling with active learning and bulk annotation features, and enables simple model implementation that can accelerate training time.

Much of what Discovery does is improve human performance.

Dan MillerAnalyst, Opus Research

Meanwhile, Dan Miller, an analyst at Opus Research, said that what the new technologies in Watson Discovery make possible in the past required a lot of human effort and manpower.

“Over the years, IBM has accelerated the path from research to product to people using the product to the needs they discover,” Miller said. “A lot of what Discovery does is improve human performance.”

Targeted highly regulated industries

The new Watson Discovery features will help clients in the financial, insurance and legal industries improve customer service and see through complex documents.

“They’re going into some of the most regulated industries, and they’re well aware of that,” Miller said.

An IBM customer using new NLP features in Watson Discovery
IBM targets highly regulated industries, including the financial, insurance and legal sectors, with new NLP features in Watson Discovery.

As these industries are highly regulated, compliance (both in terms of ethics and privacy) is important, Miller continued.

“Now it is up to solution providers like IBM to make sure they do not violate any ethical rules … [and] someone’s privacy, “Miller said.

He said IBM has stood for compliance and entered industry-specific rules into Watson Discovery.

For example, if you work with Watson Discovery to get an answer about an investment, for example, the system will show privacy rules.

And in a legal case where some information needs to be edited, Discovery contains enough intelligence to answer the question while blocking the edited information, Miller said.

“It’s a pretty complex offer,” he continued. “It seems like a lot of thought has gone into it, and it’s definitely moving the ball forward in terms of accurate research by using virtual assistants to answer questions more accurately.”

Leave a Comment