The Identify Key Phrases node allows you to identify specific words and phrases in an input string field. A 'key phrase' is a string containing a noun; this could be a single word, or a modifier and a noun. The node uses the Amazon Comprehend API to identify key phrases in incoming text fields.
The node outputs all input fields, and one new array field called 'IdentifiedPhrases'. You can click a row in the IdentifiedPhrases column to show the following information:
- BeginOffset - The position of the first character of the phrase in the string.
- EndOffset - The position of the last character of the phrase in the string.
- Score - The estimated accuracy of the analysis.
- Text - The detected text entity.
Example
Input text: It was a big house, with large windows in the countryside.
| BeginOffset | EndOffset | Score | Text |
|---|---|---|---|
| 9 | 17 | 0.9974726 | big house |
| 25 | 37 | 0.99040056 | large windows |
| 46 | 56 | 0.9912345 | countryside |
For more information, please see the Amazon Comprehend documentation, for example https://docs.aws.amazon.com/comprehend/latest/dg/how-key-phrases.html
Properties
Display Name
Specify the name of the node that is displayed on the Analysis Designer canvas.
The default value is Identify Key Fields.
Identify Key Phrases in
Select an input field to analyze.
Language
Select the language of the text to analyze.
The default value is English.