About Multi-Channel Text Analytics
The Text Analytics extracts themes and named entities from the indexed documents and stores this metadata in the index. Themes are noun phrases extracted from full text based on linguistic analysis. A noun phrase is a sequence that can be replaced by a noun or pronoun. Named Entities are categories of information like people, companies, products, or places. Themes and named entities together with other metadata feed facets and widgets that present consolidated information.
Example: In a high technology enterprise, you can use Multi-Channel Text Analytics to find and correlate knowledge base and related contents as in the following role-based insight console for assisted service.
|Facet showing tags entered by end-users|
|Facet showing themes from the result set|
|Facet showing product named entities from the result set|
|Mini-results showing recommended documents based on overall context|
|Controls to dynamically tag and link knowledge document|
|Widget showing recommended experts correlated using text analytics from result set and all enterprise content|
|Widget showing resolved tickets with similar symptoms|
|Widget showing related community content recommended through text analytics|
|Widget showing related social content recommended through text analytics|
On the back-end side, the text analytics process collects the necessary metadata in the documents. The Coveo text analytics package contains preset filter, extractor, or normalizer pipeline stages that you can assemble to create a text analytics pipeline. Developers can also use the Text Analytics framework to build custom pipeline stages.
Note: Contact the Coveo Professional Services for more information and assistance for the implementation of Multi-Channel Text Analytics.