Entity Extraction
  • 30 Oct 2024
  • 1 Minute to read
  • Dark
    Light
  • PDF

Entity Extraction

  • Dark
    Light
  • PDF

Article summary

Entity Extraction applies statistical modeling to pick up text from documents such as emails, financial figures, PII, etc. and lists as metadata fields. These fields can be viewed on the Dashboard and gives users the ability to create queries from the terms that were gathered during the analysis phase. Below is an example of faceted lists resulting from the metadata fields, "Entity_Money" and "Entity_Religion."

You have the option of enabling Entity Extraction during the Brainspace ingestion process.

  • Entity Extraction will automatically identify and extract Entities from the text of your documents regardless of document type as long as text exists.

  • Brainspace extracts and presents a standard list of Entities within the Facets drop down menu on the Dashboard. Each Entity has a set of unique values identified during ingestion. Each Entity Value will have corresponding document count which represents the number of documents that includes that specific Entity Value.

  • An Entity Value is specific Entity. Example for Entity_Email a specific Entity Value might be [email protected].

The standard list of extracted Entities include:

Entity Field Name (Facets List)

Entity_Credit_Card_Num

Entity_Email

Entity_Location

Entity_Money

Entity_Nationality

Entity_Organization

Entity_Person

Entity_Personal_ID_Num

Entity_Phone_Number

Entity_Product

Entity_Religion

Entity_Title

Entity_Url

Brainspace leverages solutions from Basis Technology for entity extraction.


ESC

Eddy AI, facilitating knowledge discovery through conversational intelligence