Overview

The Output Database is a SQL relational database created and populated by the Output Database Client. The Python Database client is used to set up the Output Database, submit reports to the NLP API and insert the structured output returned by the NLP API into the Output Database. By storing the NLP API’s structured output in a SQL database directly after processing, analysts and data scientists can immediately begin exploring and analyzing data without waiting for additional data integration or ETL steps to be performed.

This document provides an overview of the schema of the Output Database, and how it implements the underlying data model of the NLP API’s result format. In this document, we introduce the reader to the database representation of key objects in these NLP API result outputs, namely, entities, concepts, documents and relations.

The goal of this document is to enable the reader to the write SQL queries they need to retrieve and analyze in the emtellipro database.

• The content of this documentation is organized so that the data model of core features are explained first. The core capability of the NLP API is its medical NLP entity extraction service. The NLP API extracts entities and attributes from documents, and maps entities to standardized representations and semantic classifications established in several widely-used biomedical ontologies. The main tables in the Output Database are the entity and document-related tables. We recommend you familiarize yourself with these before tackling advanced data extraction features such as experiencer, medication and other features.
• While reading the documentation, it may be helpful to have an existing data set that you can use to explore the data schema and run sample SQL queries.
• To set up an Output Database and populate it with data please refer to the Output Database Client documentation.
• NLP API supports a growing number of data extraction features and ontologies. Note, that data extraction features must be enabled during data processing and may be subject to licensing restrictions. To enable a feature on a trial basis, please contact support@emtelligent.com.
• Support is available to assist you in integrating the NLP API data extraction capabilities into your data infrastructure, or with updating your current implementation. You can reach the emtelligent support team at support@emtelligent.com.