Amadeus Trip Purpose Prediction

Name: Trip Purpose Prediction API
Brand: Trip Purpose Prediction
Availability: InStock

★ Only Publicly Available OpenAPI DocumentAI/MLMl Inferenceoauth21 EndpointsREST

For Agents

Predict whether a flight itinerary is for business or leisure travel by sending origin, destination, and date pairs to a single classification endpoint that returns a labelled probability score.

Install Jentic One Beta

Connect the Trip Purpose Prediction to your agent

Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the Trip Purpose Prediction, or any other public or private API you need. You set the rules, the agent never sees your credentials, and every call is logged.

Two steps, two machines. Install the instance in a safe environment, then register your agent from wherever it runs.

Step 1: Jentic One Host machine

# On the machine that will host your Jentic One instance:
curl -fsSL https://raw.githubusercontent.com/jentic/jentic-one/main/tools/install.sh | sh

Step 2: Agent machine

# On the machine where your agent runs (keep this separate from the instance):
curl -fsSL https://raw.githubusercontent.com/jentic/jentic-one/main/tools/install.sh | sh
jentic register       # connects your agent to your Jentic One instance

Jentic One is in public beta. The setup above keeps your agent separate from the instance, which is what you want before using real credentials: an agent running as the same OS user as Jentic One can read its stored keys directly. Just evaluating? A single local install is fine to start. See the secure deployment guide for the tiers.

Capabilities

What an agent can do with Trip Purpose Prediction API.

Classify a round-trip flight itinerary as business or leisure based on city pair and dates

Return a probability score so agents can apply confidence thresholds before acting on the prediction

Use the search-date input to factor advance-purchase patterns into the trip-purpose decision

GET STARTED

Start building with Trip Purpose Prediction API

Explore with Jentic

View OpenAPI Document

Use Cases

Patterns agents use Trip Purpose Prediction API for, with concrete tasks.

★ Business or Leisure Segmentation for Marketing

Travel platforms use the Trip Purpose Prediction API to classify each flight search as business or leisure before rendering merchandising. The single /travel/predictions/trip-purpose endpoint accepts an origin, destination, departure date, return date, and search date, and returns a result label with a probability. Marketing teams use the score to decide whether to show flexible-fare upsells and lounge passes versus holiday hotel bundles, with classification typically completing in well under a second per request.

Call /travel/predictions/trip-purpose with originLocationCode=NYC, destinationLocationCode=MAD, departureDate=2026-08-01, returnDate=2026-08-08, searchDate=2026-06-09 and surface the predicted label and probability

Personalised Itinerary Content

Online travel agencies enrich each search result with a trip-purpose label and use it to choose images, copy, and ancillary upsells. Leisure-classified itineraries surface tours and excursions, while business-classified ones surface fast-track security and chauffeur transfer offers. The probability score lets the application fall back to a neutral layout when confidence is low rather than showing the wrong content.

For each itinerary in a search results array, call the trip purpose endpoint and tag the result with label business or leisure and the probability value

Travel Analytics Cohort Building

Analytics teams replay historical flight-search logs through the Trip Purpose Prediction API to build business and leisure cohorts for retention reporting, conversion analysis, and demand forecasting. Because the API requires only city pair, dates, and the original search date, it can be applied retroactively to any logged search without needing personal traveller data.

Iterate over a CSV of historical flight searches and call the trip purpose endpoint for each row, writing the predicted label and probability back to the dataset

AI Agent Trip Triage

An AI travel-planning agent invoked through Jentic uses the Trip Purpose Prediction API as a triage step before calling downstream Amadeus search and booking endpoints. The agent searches Jentic for the trip-purpose operation, loads its schema, executes a single call with the candidate itinerary, then routes business-classified trips to corporate-rate workflows and leisure-classified trips to package-deal workflows.

Search Jentic for predict trip purpose, load the trip purpose operation schema, execute it for the user's candidate itinerary, and branch to corporate or leisure workflow based on the returned label

Key Endpoints

1 endpoints — the amadeus trip purpose prediction api classifies a flight itinerary as either business or leisure travel.

METHOD

PATH

DESCRIPTION

GET

/travel/predictions/trip-purpose

Predict whether a trip is business or leisure for a given city pair and date range

GET

/travel/predictions/trip-purpose

Predict whether a trip is business or leisure for a given city pair and date range

Why Jentic?

Three things that make agents converge on Jentic-routed access.

Credential isolation

Amadeus client_id and client_secret are stored encrypted in the Jentic vault. Jentic performs the OAuth 2.0 token exchange server side and supplies a scoped bearer token to the request, so the agent never sees the raw Amadeus secret.

Intent-based discovery

Agents search Jentic with phrases like predict trip purpose or classify business or leisure flight, and Jentic returns the trip-purpose operation with its input schema, so the agent can call it without browsing Amadeus Self-Service docs.

Time to first call

Direct integration: 1 to 2 days for OAuth setup, error handling, and quota management. Through Jentic: under 30 minutes — search, load schema, execute.

FAQs

Specific to using Trip Purpose Prediction API through Jentic.

What authentication does the Trip Purpose Prediction API use?

The API is part of the Amadeus Self-Service catalogue and uses OAuth 2.0 client-credentials. You exchange a client_id and client_secret for a short-lived bearer token at the Amadeus auth endpoint, then send it on the trip purpose request. Through Jentic the client_id and client_secret are stored encrypted in the Jentic vault and the token exchange happens server side, so the agent never sees the raw secret.

Can I classify multi-city itineraries with the Trip Purpose Prediction API?

No. The /travel/predictions/trip-purpose endpoint accepts a single origin, destination, departure date, and return date, so it is designed for round-trip city pairs. For multi-leg itineraries, classify each leg separately or use the longest leg as a proxy.

What are the rate limits for the Trip Purpose Prediction API?

The Amadeus Self-Service tier limits this endpoint to 10 transactions per second per application in the test environment, with separate per-month transaction caps. Production quotas are negotiated per Amadeus contract. Treat any 429 response as a signal to back off and retry.

How do I predict trip purpose with the Trip Purpose Prediction API through Jentic?

Install the SDK with pip install jentic, then run a Jentic search for predict trip purpose, load the resulting operation schema, and execute it with originLocationCode, destinationLocationCode, departureDate, returnDate, and searchDate. Jentic returns the label and probability without you handling the OAuth token directly.

Is the Trip Purpose Prediction API free?

The Amadeus Self-Service free tier includes a monthly quota for trip purpose predictions in the test environment, suitable for prototyping. Production access is on the standard Amadeus commercial pricing and requires moving the application to the production environment.

What inputs does the Trip Purpose Prediction API actually require?

Five fields: originLocationCode and destinationLocationCode as IATA city or airport codes, departureDate and returnDate as ISO dates, and searchDate as the date the user performed the search. The model uses the gap between searchDate and departureDate as a key signal, so do not substitute the current date.