citlyze docs
API

Sentiment

Net Sentiment Score aggregates per brand, engine, and measurement window.

GET /api/v1/sentiment returns Brand Perception aggregates: for each measurement window, brand, and engine, the sentiment counts across that window's answers and the resulting Net Sentiment Score. The attribute side of perception lives in attributes and attribute mentions.

Filters

ParameterMeaning
measurement_window_idRestrict to one window.
brand_idRestrict to one brand (yours or a competitor).
engine_idRestrict to one engine.

Standard parameters (limit, offset, format=csv) work as described in Pagination, filtering, and errors.

Key fields

  • measurement_window_id: joinable to measurement windows
  • brand_id: the brand the row scores, joinable to brands
  • engine_id: the engine whose answers were classified
  • positive_count / neutral_count / negative_count: classified brand mentions in the window
  • caveat_count: mentions that praise with a reservation
  • mention_count: total classified mentions behind the row
  • nss: Net Sentiment Score, from -100 to 100, mention-weighted
  • avg_confidence: average classification confidence, 0 to 1

Rows are ordered by measurement_window_id descending with brand_id and engine_id as tiebreaks, so offset pagination is stable.

Example

curl -H "Authorization: Bearer $AEO_API_KEY" \
  "https://app.citlyze.com/api/v1/sentiment?brand_id=$BRAND_ID"

Use this resource to trend how AI answers talk about each brand over time and to compare tone across engines and against competitors.

本頁內容