| Excerpt |
|---|
Any items that are added to Squirro go through the Squirro Pipeline. In that Pipeline, the items are normalized and enriched, before they are written into the index. This process is called Data Enrichment and is highly configurable. |
Table of Contents
| Table of Contents | ||||
|---|---|---|---|---|
|
Webinar
Working with enrichments is covered in the Enrichment Webinar.
Types of Enrichments
Four different type of enrichments are available:
- Built-in enrichments which can be configured on a per-source basis.
- Search Tagging can add keywords to items based on their matching a pre-defined search query including Smart Filters.
- Known Entity Extraction which can link content to entities extracted from a CRM or other internal systems.
- Squirro Script is a scripting language embedded in the Squirro pipeline.
- Pipelets are plugins for the pipeline that can be added to the Squirro platform by 3rd party developers.
Configuring Enrichments
Enrichments can be configured on two levels: per project and per source.
Project enrichments
The user interface provides access to project enrichments in the Enrich tab. The Search Tagging and all available Pipelets are configurable through this user interface by project administrators.
The same functionality is also available through the API by using the Enrichments resources.
Source enrichments
The Built-in enrichments and the Pipelets can be configured for individual sources. This is done with the processing config, which can be provided when creating or updating subscriptions (see the Subscriptions resource).
Running Enrichments
Enrichments are automatically executed by the pipeline when data is ingested.
...
The Pipeline has a number of different steps available. These are documented here in their various sections. |
Table of Contents
| Child pages (Children Display) |
|---|