Saga Natural Language Understanding (NLU) Framework
A new R&D initiative to offer a scalable framework for creating and maintaining natural language models for user interaction and document understanding
Nearly 80 percent of enterprise content is unstructured. It consists of fast-growing human-generated content, including memos, emails, text documents, research and legal reports, voice recordings, videos, social media posts, etc. Unlike structured content (tables, forms, log files), it is difficult to search for, let alone analyze, meaningful information from unstructured, natural language content.
As enterprises increasingly become insight-driven, they are seeking to leverage the vast unstructured data to improve business operations and accelerate speed to outcomes. But existing natural language processing and understanding (NLP/NLU) technologies are not fulfilling enterprise demands—they are too narrow, too generic, or too costly to develop, deploy, and maintain.
Saga Natural Language Understanding (NLU) was developed to provide a scalable framework that fills the gaps in existing NLP/NLU technologies. See more.
View TranscriptA new R&D initiative, Saga NLU was developed to provide a scalable, easy-to-use framework that fills the gaps in existing NLP/NLU technologies. It features:
Business-friendly, code-free user interfaces for creating and maintaining language models for a wide range of enterprise NLP/NLU use cases
Automated NLU pipeline construction and management which substantially reduces development costs and provides predictable cost of ownership
Flexible, scalable deployments on-premises or in the cloud
State-of-the-art handling of language ambiguity
Integrated machine learning with many prepackaged models available out-of-the-box
Extensible with machine learning and knowledge graph technologies
Saga can be used as a standalone NLU framework or together with our range of technology assets designed to optimize the performance of search, analytics, and NLP applications.
NLU capabilities are powered by both Patterns Matching (for precision and ease of editing) and Machine Learning (for broad coverage and automatic learning).
Patterns are simple to understand, accurate, quick to show value, and work best when no training data is available. NLP output with business object IDs can be easily integrated into business actions.
The process of testing and deploying Machine Learning and language models is easily done and managed by non-data scientists as it does not require coding.
Non-data scientists can perform 95 percent of the NLP/NLU work, providing “ready-to-go” data for data scientists to focus on creating better models.
Enterprises have full control over solution performance, deployment, and cost predictability.
Multiple NLP development efforts, algorithms, and language resources are managed and coordinated centrally within the Saga framework.
Saga NLU is flexible, easy-to-use, and scalable to support a variety of enterprise NLP/NLU needs. Example enterprise use cases include:
Contact us to request a Saga demo and learn more about how Saga can support your NLU use cases.