SmartHub Integration – Rasa Stack

As new capabilities in natural language processing, machine learning, and cognitive computing continue to arrive, adopters face challenges with the “cloud first” or “cloud only” approaches of Microsoft and Google. For many environments, this external integration and data exchange point is not something that can be supported. To address this, BA Insight has developed integrations with The Rasa Stack, an open source natural language processing and dialogue management tool that is supported by thousands of community members.

BA Insight has integrations with a number of AI frameworks, so it has been natural to apply our best-of-breed approach and harness Rasa’s open source capabilities to create intelligent search.

Rasa Artificial Intelligence IconRasa provides complete capabilities, all of which are integrated into SmartHub’s Natural Language Query Offering:

  • Contextual Conversations: Complete support for context and dialogue management, allowing real time conversations and true AI assistance capabilities.
  • Recognize Intents: Understand the meaning of a sentence in dozens of natural languages. For advanced use cases, train custom language models for even better performance – all using open source NLU.
  • Extract Entities: Get valuable, structured data like dates, addresses and phone numbers from free text. Save time and reduce training data burden by relying on built-in, pre-trained entities using Rasa NLU.
  • Complete Data Control: Data is your competitive edge. Don’t send your valuable training data to third party APIs. Remain independent: deploy on-prem or in a private cloud.

Our modular approach has always allowed our customers to rapidly assemble powerful search solutions. We are now able to provide dramatically better search by The Rasa Stack.

Better Search – using machine learning and cognitive APIs we can enhance familiar search capabilities including:

  • Semantic Data Enrichment:  Information is much more powerful when it is cross-related and there is a wealth of linked and semantic data that can be used to make much more powerful search and discovery experiences
  • Relevanceis improved by using machine learning to adjust based on successful searches, personal preferences, and other data collected via our SmartHub Analytics product
  • Recommendations: personalized search, recommendations, suggestions, and adaptive UIs are all made possible by using advanced machine learning

New Types of Search and Discovery – by using AI we can provide completely new types of interactions with your data, such as:

  • Multi-media search:  image, speech, and video understanding has also made huge advances due to Machine Learning and Natural Language Processing. There is a lot of image and video content inside organizations that has traditionally been out of range for search.
  • Multi-lingual search using machine translation: Microsoft has made huge advances in this area, and machine translation is opening a much wider audience for content and collaboration.
  • Spoken Language search provides a more natural, interactive user experience.  It also allows hands-free interactions – from mobile phones, in laboratory and factory environments, and in augmented reality applications.
  • Search Bots provide information in chat, email, and social media applications including Teams, Yammer, and Skype.

We have already built a number of these capabilities into our products and are actively building out more of them.

Contact us today to learn more.