cheapair semantic search flight booking natural query OTA
664 days ago
 

CheapAir unveils semantic search for booking flights with natural language queries

Aiming to make travel search more intelligent, CheapAir.com has begun beta-testing Easy Search, a semantic search tool that allows users to pose natural language flight queries instead of choosing parameters in forms, drop-down menus, and calendars.

CheapAir claims to be the first online travel agency (OTA) in the US to embrace semantic search.

The tool can be found on CheapAir’s homepage, by clicking the Easy Search tab, which is the fourth option on the top navigation bar.

cheapair semantic search flight booking natural query OTA
Passes the sniff test

CheapAir’s search box prompts users with a leading question: “Tell us the cities and dates of your trip.”

When I tested the query “Chicago to LA on Oct 5th, back on the 10th,” Easy Search fetched adequate answers.

My query for “LGW to TYO 26oct 31st oct” was also correctly understood as “London, United Kingdom to Tokyo, Japan, October 26, 2012 – October 31, 2012.” I could then click with a button for how many tickets I needed and CheapAir retrieved fare listings.

A narrow range of possible queries

CheapAir directs customers to only make queries with cities and dates.

The site’s search box accepts city and airport codes and allows users to enter dates in any format.

A few sample searches are listed as examples, guiding users in how to use the tool and boosting the chances customers will see helpful results.

Semantic search algorithms have additional considerations of grammar and syntax along with the statistical ones of word sequence, which rely on programmers to attach extra, machine-readable code to every page of results.

Where the terminology is limited, semantic search can produce good enough results.

Yet the more open-ended a query, the greater the chance of receiving a faulty answer, as Apple’s Siri often illustrates in a different context.

Earlier this year, iXiGO.com, an Indian travel search engine, launched a natural language flight search tool. But its search results have not been as robust as CheapAir’s.

For instance, a recent search on iXiGO — “chicago denver by air” — returned the information: “Chicago (Illinois) to Denver (Colorado) does not have flight connectivity.”

A big stride for a “second-tier, but important” OTA

The move is a bold experiment for CheapAir, which has 50 staffers and says it “crunches more than 25 million fares a day.”

But the Calabasas, Calif., company has an Achilles’ heel.

In 2011, Delta Air Lines, which serves more than 160 million customers a year, pulled its fares from CheapAir, leaving a big hole in the OTA’s search results.

Southwest Airlines, which carries more passengers domestically than any other US airline, also does not provide fares for the search engine, thought that is the carrier’s policy toward all OTAs.

Perhaps natural language queries will draw enough buzz to draw the consumer volume that forces Delta and other airlines to take it more seriously.

The site can claim a first for semantic search in the US, getting a jump on Hopper, a start-up that will likely specialise in semantic search for travel.

 
 
Sean O'Neill

About the Writer :: Sean O'Neill

Sean O’Neill is a New Jersey-based reporter for Tnooz. He's also a regular contributor to BBC Travel.

Follow him on Twitter, Google+, and his personal site .

 

Comments

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

  1. Chris T

    My understanding of semantic search is respectfully different to what transpire here. A semantic search would respond to questions such as “cheapest air fare LHR to JFK next weekend”. Anything different would be classified as organized results based on set search terms.

     
  2. Sean O'Neill

    Sean O'Neill

    Hi, Evan,
    I asked but I’m afraid they didn’t say. Will ask again nicely! :-)
    Best,
    Sean

     
    • Jeff Klee

      Evan/Sean:

      It’s our own technology. We’ve built it in-house. This is just a first crack at it and, frankly, its far from perfect. But we’re improving it pretty rapidly and we’re pretty excited about all of the potential applications.

      - Jeff

       
      • Sean O'Neill

        Sean O'Neill

        Jeff,
        Thanks for chiming in to the conversation!

        It sounds like a basic parser engine, but an effective one at what it does.

        Sean

         
  3. Evan Konwiser

    Have they disclosed what technology they are using for the parsing?

    I keep waiting for Evature tech to make it into more travel brands. Semantic search might not be a save-all, but it sure will make things better.

     
 
 

Newsletter Subscription

Please subscribe now to Tnooz’s FREE daily newsletter.

This lively package of news and information from Tnooz’s web site provides a convenient digest of what’s happening in technology that drives the global travel, tourism and hospitality market.

  • Cancel