Technology
You need information? Go ahead and ask. In your own words.
Semantra resolves the linguistic mismatches between common business language and computer language, enabling every employee in the organization to turn critical questions into precise answers and actionable information.
The principle parts of the solution include: 1) a semantic engine that parses natural language, 2) a collection of ontologies and business rules that provide context, and 3) a dynamic query generator that creates the appropriate SQL (structured query language) command to be executed against a targeted database. These foundational elements form a “semantic layer” between knowledge workers and enterprise applications with their relevant corporate data source(s).
Semantra’s NLP (Natural Language Processor) is the core semantic engine responsible for parsing users’ common-language requests. The user submits the inquiry by typing into a familiar search box. The NLP breaks down the sentence structure, interprets the grammar and phrases, deciphers synonyms and parts-of-speech elements, and even resolves misspellings on the fly. Should any semantic mismatches or ambiguities still exist, the NLP assists the user in clarifying the request.
From inquiry to result:
Once parsing is complete, the inquiry is mapped against the OntoloNet™, a hierarchical repository of business concepts, terminology and business rules that collectively form the backbone of Semantra’s technology. This ever-expanding and adaptable knowledge base allows most any business to quickly bootstrap semantics into their enterprise while inheriting any shared (non-proprietary) concepts, business rules, jargon, and acronyms from Semantra’s vertical industry ontologies. These concepts and terms, along with target database metadata, are captured during Semantra’s pre-deployment process, known as “semantification.”
Once the inquiry is fully understood, the final step is to dynamically convert the request into the appropriate query command (SQL or related format) which is executed against the structured enterprise database(s). Within seconds of the initial request, the results of the query are displayed in the user’s browser, in the selected graphical or tabular format. Additionally, query results can be exported to third-party applications such as business intelligence systems, report beautifiers, charting or mapping tools, or spreadsheets.