Skip to main content

Text Summary under Text Mining_256baa0.png Text Summary

The Text Summary tool summarizes bodies of text. Use the Text Summary tool to better understand bodies of text or as an aid for further downstream tasks that involve large bodies of text. The tool leverages the text summarization capabilities in Summa - Textrank.

Alteryx Intelligence Suite Required

This tool is part of Alteryx Intelligence Suite. Intelligence Suite requires a separate license and add-on installer to Designer. After you install Designer, install Intelligence Suite and start your free trial.

Language Support

The Text Summary tool supports English, French, German, Italian, Portuguese, and Spanish.

Tool Components

The Text Summary tool has 2 anchors (1 input and 1 output):

  • Input anchor: Use the input anchor to connect the text data you want to summarize.

  • Output anchor: Use the output anchor to pass the summarized text downstream.

Configure the Tool

  1. Add a Text Summary tool to the canvas.

  2. Use the input anchor to connect the Text Summary tool to the text data you want to use in the workflow.

    1. To bring text data into Designer, use an Input Data or Text Input tool.

    2. We recommend that you remove special markings between sentences. For example, remove citation numbers at the end of sentences such as [number], (number), and number. You can remove text like this with the RegEx Tool.

  3. Select the input text Language. If your text contains multiple languages, select the most prevalent language.

  4. Select the Column with Text you want to summarize.

  5. Enter the number of sentences for your Summary Length. The text summary must have fewer sentences than your text.

  6. Select the Summary Method.

  7. Run the workflow.

Summary Method

The text summary output might vary based on the algorithm or model you choose. We recommend TextRank by default. This list includes the current summary methods we support:

  • TextRank: A graph-based algorithm that creates a text summary from important sentences. TextRank uses the extractive summary methodology. This methodology identifies important sentences and returns them verbatim in order of importance. The summary might not be as articulate as a human-generated summary. TextRank performs best on blocks of text. Remove headings, footnotes, and other parts of the document that aren't part of the main body of the text. For text with multiple languages, TextRank works best with the most prevalent language.

Output

The output includes 1 column that contains your text summary. Connect the Report Text tool to compare your original text to your text summary.