Schlagwort-Archive: travel patterns

Travel Blogging

„While travel blogging is a relatively young phenomenon, it has already evolved into a mature and sophisticated business model, with participants on both sides working hard to protect and promote their brands.

Those on the industry side say there’s tangible commercial benefit, provided influencers are carefully vetted.
„If people are actively liking and commenting on influencers‘ posts, it shows they’re getting inspired by the destination,“ Keiko Mastura, PR specialist at the Japan National Tourism Organization, tells CNN Travel.
„We monitor comments and note when users tag other accounts or comment about the destination, suggesting they’re adding it to their virtual travel bucket lists. Someone is influential if they have above a 3.5% engagement rate.“
For some tourism outlets, bloggers offer a way to promote products that might be overlooked by more conventional channels. Even those with just 40,000 followers can make a difference.
Kimron Corion, communications manager of Grenada’s Tourism Authority, says his organization has „had a lot of success engaging with micro-influencers who exposed some of our more niche offerings effectively.“
Such engagement doesn’t come cheap though.“

That means extra pressure in finding the right influencer to convey the relevant message — particularly when the aim is to deliver real-time social media exposure.
„We analyze each profile to make sure they’re an appropriate fit,“ says Florencia Grossi, director of international promotion for Visit Argentina. „We look for content with dynamic and interesting stories that invites followers to live the experience.“
One challenge is weeding out genuine influencers from the fake, a job that’s typically done by manually scrutinizing audience feedback for responses that betray automated followers. Bogus bloggers are another reason the market is becoming increasingly wary.“

Airbnb Is Quietly Building the Smartest Travel Agent of All Time

Airbnb-14

Airbnb overhauled its logo, its website, and its mobile app this morning. But there’s something deeper going on with the sharing economy’s most popular travel site.

Under the covers, Airbnb has quietly begun an ambitious effort to painstakingly mine the treasure trove of data contained in the site’s customer reviews and host descriptions to create a smarter way of traveling. It turns outs Airbnb is more than a travel website—it’s a stealth big data company.

“For a long time now, Airbnb has been an awesome place to go if you know where you’re going and you know when you’re going,” says Mike Curtis, Airbnb’s vice president of engineering. “But we realized that we have all of this data that other people don’t have. We have travel patterns. We have the reviews. We have the descriptions of the listings. We know a lot about neighborhoods that we can infer from the text in there.”

To do this, the company has formed an eight-person Discovery team. Their mission? To build language processing software that mines Airbnb’s data and figures out what’s really happening out there in the travel world. In other words, Airbnb is building a kind of omniscient, machine-powered travel agent of the future.

‘WE REALIZED THAT WE HAVE ALL OF THIS DATA THAT OTHER PEOPLE DON’T HAVE. WE HAVE TRAVEL PATTERNS, WE HAVE THE REVIEWS, WE HAVE THE DESCRIPTIONS OF THE LISTINGS.’

You can see the early hints of this in the new recommendations that debut on the site today. Airbnb figures out where you’re from, and then drops you a few travel ideas. “We try to figure out exactly where you are and who the people are around you and where they like to travel,” says Surabhi Gupta, an engineer on the Discovery team.

If you’re booking from Knoxville, Tennessee, for example, there’s a pretty good chance you’ll want to take in the sights in Washington, DC. If you’re from the San Francisco or Brooklyn, you may very well be looking for a booking in the same city (folks in these places are more likely to be using Airbnb to book accommodations for friends or relatives).

 

Airbnb-Discovery-14

 

The Discovery team figures this out by extracting interesting words from the site’s reviews and descriptions. An open-source tool called the Stanford Part of Speech Tagger comes in handy for this. It then uses custom-build algorithms to assign 150 different attributes—beaches, hiking, sunsets, and so on—to different locations.

What you see on the homepage is a start, but Airbnb wants to get to the point where it can give very specific recommendations based on who you are, not just where you live. “A lot of what we’re doing is the foundational work for user-level personalization,” says Lu Cheng, another Discovery team engineer. That means, in a few years, you may very well be using Airbnb to not only book your next vacation, but to figure out where the heck you want to go.

 

Source: http://www.wired.com/2014/07/airbnb_recommendations/