HPE IDOL Machine Learning Engine Adds Natural Language Processing

Print Friendly, PDF & Email

hpeToday Hewlett Packard Enterprise announced a new release of HPE IDOL, the company’s flagship unstructured data analytics engine featuring advanced Natural Language Question Answering. The new version of HPE IDOL leverages advanced machine learning functionality to improve the effectiveness and contextual accuracy of human interactions with computers.

Among the biggest challenges facing organizations trying to leverage Big Data is providing answers to users’ questions in a natural, effective manner  without cumbersome user interfaces or extensive training. Interactive voice assistants and online chatbots have recently simplified this process for consumers, however developers have had a difficult time adapting this approach to enterprise-class tasks due to the complexity and context of the questions, trustworthiness of the source, specificity of the information needed and accuracy of the answer.

HPE Natural Language Question Answering deciphers the intent of a question and  provides an answer or initates an action drawing from an organization’s own structured and unstructured data assets in addition to available public data sources to provide actionable, trusted answers and business critical responses.

Building on HPE IDOL’s history of delivering industry-leading analytics engineered for human data, IDOL Natural Language Question Answering is the industry’s first comprehensive approach to delivering enterprise class answers,” said Sean Blanchflower, vice president of engineering, Big Data Platform, Hewlett Packard Enterprise. “Designed to meet the demanding needs of data-driven enterprises, this new, language-independent capability can enhance applications with machine learning powered natural language exchange.”

To deliver natural language based systems that meet the demanding needs of knowledge workers, it is critical to deliver answers that are accurate, relevant and trusted. HPE IDOL Natural Language Question Answering is a core feature of the new HPE IDOL 11.2 software release that features four key capabilities for natural language processing for the enterprise:

  • IDOL Answer Bank: Delivers precise, curated responses to predetermined reference questions. For example, IDOL Answer Bank could be programmed to deliver detailed step-by-step instructions for configuring a specific model of a mobile phone.
  • IDOL Fact Bank: Provides fact-based answers, such as the price of a stock on a specific date, by querying existing structured enterprise data sources or by employing sophisticated table extraction methods to extract precise data from unstructured data sources such as a company’s annual report.
  • IDOL Passage Extract: Designed to provide text-based overview information on topics, people or events by analyzing freeform text data sources for contextually relevant information to provide a summary text passage response. This feature could be used to provide an overview of a recently enacted financial services regulation or a news event.
  • IDOL Answer Server: Analyzes questions and available data sources to determine how to best provide an optimal answer by harnessing the IDOL Answer Bank, Fact Bank, and Passage Extraction natural language answer engines.

HPE IDOL Natural Language Question Answering is architecturally designed to be language independent and can be seamlessly integrated with most third party dialogue flow systems, such as the popular NADIA open source natural language system, to create natural, back and forth conversational exchanges.

Transforming access to contextually relevant and critical information to improve patient care is central to our mission,” Dan Schlake, Strategic Accounts Director, ChartMaxx, Quest Diagnostics. “With HPE IDOL Natural Language Question Answering as the next addition to ChartMaxx Deep Search, we can see a path forward to address the critical last mile of computing in which sophisticated information exchange happens between humans and machines in a natural and intuitive fashion.” 

Sign up for our insideHPC Newsletter