Skip to content

Tutorials Quotation

docs-mirror edited this page Aug 4, 2026 · 3 revisions

← Back to tutorial index

Quotation Extraction tutorial

Quotation Extraction identifies quoted speech, speakers, and speech verbs in English news-style text. The built-in rule-based engine is based on the Gender Gap Tracker work from Simon Fraser University's Discourse Processing Lab.

The rules were developed for Canadian news. Results may be less accurate for social media, fiction, historical documents, or other English varieties. Review a representative sample before drawing conclusions from the output.

Parameter panel

Step 1 — Select your data

Add one Data Block and choose the source text column. A fresh selector uses the Data Block's saved Document Column Preference when available. The Analysis records the exact Data Block and column used for the run.

Step 2 — Choose the engine

The engine is an Analysis parameter in the Quotation panel:

  • Built-in is the default and runs the bundled local quotation engine. It needs no separate service, URL, or user configuration.
  • Remote sends the work to an endpoint configured by the deployment operator. Enter the operator-provided Engine id. Wordflow does not accept arbitrary service URLs from the browser, and an unknown ID is rejected.

Use Remote only when the administrator of your Wordflow deployment has given you a valid engine ID and its data-handling policy is appropriate for the text. The immutable Analysis request stores Built-in or the selected remote engine ID, never a host URL.

Step 3 — Set display context

Context length (words per side) controls how much source text the Result table displays around the highlighted quotation spans. It is a presentation setting, not an extraction-engine parameter.

  • Default: 5 words per side.
  • Range: 0–2000.
  • Use 0 to keep the display close to the extracted speaker, quote, and verb.

Step 4 — Preview

Click Preview to create a durable Quotation Preview Analysis. If the Data Block, text column, or engine changes afterward, click Update Preview to replace it with a new immutable request.

The successful Preview Analysis keeps its retained input snapshot. Each page and sort request recomputes a fresh projection from that snapshot, not the current mutable Data Block. Preview pages are not cached.

Result panel

The table pages through source documents and omits documents with no extracted quotation. A source document can contribute several quotation rows.

Colour Entity Meaning
Blue Speaker The person attributed as speaking
Green Quote The quoted text
Violet Verb The speech verb, such as said or argued

Click a row to inspect the full source document. The metadata selector can add source fields and generated quotation fields to the table. The virtual QUOTE_extraction document header sorts by the Analysis's selected source text column. Other source metadata headers remain sortable; generated quotation headers are display-only because they are produced after source paging.

Changing the page, documents-per-batch value, or sort order recomputes another projection of the same Preview Analysis. It does not mutate that Analysis.

Run All and Review

Click Run All at any time to submit an independent Run All Analysis that retains a complete table Result from its own immutable snapshot. Later source edits do not alter that Analysis's meaning, and Run All does not add a Data Block to the Workspace. After success, Review reads the immutable Result. Page by Documents shows the highlighted reading view, while Page by Matches shows one raw extract per row with scalar QUOTE_* fields. Changing the paging unit returns to page 1. Review does not show the Preview page summary.

Use Add to Workspace to publish selected Result columns as a Derived Data Block. The document column is required, metadata columns start unselected, and analysis columns start selected.

Clear results

The Tab retains its Analysis forest across navigation and Workspace reopen. Clear Results removes the complete forest, including after failure or cancellation.

Troubleshooting

Symptom Likely cause What to try
Remote engine is rejected The ID is empty or not configured by the operator Use Built-in or ask the deployment administrator for a valid ID
No quotations are shown on one page The current source-document batch has no extracted quote Continue to the next page
Precision is low The text differs from the news style targeted by the rules Review the disclaimer and validate a representative sample
A generated header does not sort Generated quote fields are computed after source paging Sort by the document header or source metadata
Preview does not reflect a later Data Block edit You are viewing the historical Preview snapshot Click Update Preview to capture the changed source deliberately

Quick-reference defaults

Setting Default Notes
Engine Built-in Remote requires an operator-configured engine ID
Context length 5 words per side Display-only, range 0–2000
Preview source Immutable Preview snapshot Every page and sort request is recomputed

Practice exercise

  1. Select a news Data Block and Preview with the built-in engine.
  2. Inspect highlighted speaker, quote, and verb spans in several rows.
  3. Change the display context length.
  4. Sort by the virtual document header and a source metadata column.
  5. Run All, inspect Review, and use Add to Workspace if you need a Derived Data Block.

← Back to tutorial index

Clone this wiki locally