ATPCO tackles challenge of matching AI travel requests to airline products
A two-week proof of concept tested how structured airline data could help turn conversational travel requests into more specific shopping results., CredSpark is a powerful, interactive content platform that helps organizations maximize the
Source
What It Says
Airline tariff publishing company ATPCO has developed a new technical approach to bridge the gap between conversational artificial intelligence (AI) and structured travel inventory. The company completed a two-week proof of concept designed to translate unstructured, natural language travel queries into the highly structured data formats required by airline reservation systems.
According to Anand Mishra, VP of Technology at ATPCO, modern AI assistants excel at interpreting natural language but struggle to map those requests to specific inventory attributes. The initiative focuses on translating subjective traveler desires, such as a request for a "comfortable seat," into standardized, searchable airline product features.
Why It Matters
As travelers increasingly turn to conversational AI search tools to plan trips, the travel industry faces a major data translation bottleneck. Traditional booking engines rely on rigid, structured parameters (such as dates, airport codes, and cabin classes).
If distribution platforms cannot accurately map conversational, intent-based queries to specific product attributes, AI search tools will fail to deliver accurate booking options. Solving this translation layer is critical for enabling true natural-language commerce across all travel sectors.
Useful Signals
- The Semantic Gap: AI can easily parse the intent behind a user's conversational query, but legacy distribution systems cannot process these requests without a translation layer.
- Standardization of Attributes: To make conversational search work, subjective terms (e.g., "comfortable," "family-friendly," "quiet") must be mapped to concrete, structured data points.
- Rapid Prototyping: ATPCO's two-week proof of concept indicates that major distribution players are moving quickly to establish the infrastructure needed for AI-driven search.
STR Tech Report Take
While ATPCO’s initiative is focused on the aviation sector, the underlying challenge is identical to the one facing the short-term rental (STR) industry. Guests frequently search for accommodations using subjective language—asking for a "romantic getaway," a "good remote work setup," or a "child-safe backyard."
Currently, online travel agencies (OTAs) and property management systems (PMS) rely on rigid amenity checklists. For STR technology vendors, the race is on to build similar semantic translation layers. Property managers who enrich their listing content with structured, machine-readable metadata will be the first to benefit as search engines transition from keyword matching to AI-driven semantic matching.
Original Source
PhocusWire: ATPCO tackles challenge of matching AI travel requests to airline products
Get more insights like this
Weekly STR tech updates. No spam.