Skip to end of banner
Go to start of banner

Insights Connectors

Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Version History

Version 1 Next »

This feature is currently in the proof-of-concept phase

Introduction

Our strategy is to make the customer’s data more alive by offering insights into their data. For example, offer a summary of a document, a list of named entities, and a sentiment analysis. The Consono case, which offers a wide range of analyses, fits in this case. Additionally, we want these insights to be integrated into the front end, specifically in the context of a single object.

Connector

Schematic

To realize this, we will create a new category of connectors called “Insight”. In the sketch below there are two flows

  1. In the ingest workflow, that requests the insights connector to analyze the object and (a)synchronously waits for the analysis to complete (drawn as the steps A-B-C-D)

  2. The end user that requests and receives insights into their document (drawn as the steps 1-2-…-6)

Connector API contract

The connector of type insights will have to implement the following methods

Method

Goal

Input

Output

POST /analyze

Prepare the analysis during the ingest because it can take minutes to complete

  • Metadata

  • Including the PathToPreview URL

None

GET /insights

Provide an overview of the available insights into the object, such that these can be combined over multiple insight connectors

  • Metadata

  • List of available insights

GET /insights/:insightId

Provide the data of a specific insight, such that the front-end can render it, possibly with the aid of a Javascript library

  • Metadata of the object

  • ID of the particular insight

  • The data of a particular insight

 

  • No labels