Study Details

Explore the world via Google Street View. MapCrunch teleports you to a random place in the world. Discover the vast array of imagery captured by Google in 50 countries, featuring spectacular scenery, magical moments and the utterly unexplainable.

• An online survey was conducted with consumers who had searched online for a restaurant or hotel within the previous 30 days (using any device).
• A total sample of 1201 respondents wascollected between July 21st and August 2nd, 2015.
• Different versions of a local business listing (one for a hotel, the other for a restaurant) were created to include basic info, photos and/or a virtual tour (the same media were used to create each version).
• Respondents were asked to view one of the different versions for each business (in a randomized order). Each respondent viewed:
  • 1 hotel listing + 1 restaurant listing; and
  • 1 listing with a virtual tour + 1 listing without a virtual tour.
• Note: Presentation of alternate listings were randomized evenly and consistently throughout the field.
• After viewing each listing, respondents were instructed to rate their interest in booking a reservation.
• Ratings for alternate listings (with a tour versus without) were aggregated to understand the impact of virtual tours on interest in the featured business.

Download overview [PDF]

Each year, the U.S. Census Bureau spends $1 billion surveying the population. These surveys are designed to tease apart the demographic makeup of the country by asking a representative group of people about their race, gender, education, occupation, and so on. This is an important exercise because it gives a crucial bird’s-eye view of the population and how it is changing.But it is not easy.

For a start, the data is relatively large scale—the Census Bureau’s main survey, the American Community Survey, gives results for all cities and counties with a population greater than 65,000. Smaller areas don’t figure.What’s more, surveying the population is a time-consuming exercise; so much so that some data can be five years old by the time it is published. And because other data is much newer, comparisons can be problematic.So demographers would dearly love to have a faster, more efficient, and higher-resolution way of studying the population.Enter Timnit Gebru at Stanford University and a few pals, who have used Google Street View images to make remarkably accurate assessments of the demographic breakdown in a wide range of U.S. Their technique shows how demographers can collect reliable data in an entirely new way that complements and improves on current methods.Gebru and co begin with 50 million Street View images gathered by Google’s cars in 200 American cities. The team believes the type of car people own is a strong indicator of their race, income, education levels, occupation, and so on.So Gebru trained a set of machine-vision algorithms to recognize cars in these images and then classify each in one of 2,657 different categories that depend not only on the make and model of the car but also its age.

This gives an accurate idea of the car’s worth.In total, they classified some 22 million vehicles, about a third of all vehicles on the road in these cities. And it took their machine about two weeks to do the task at a rate of about 0.2 seconds per vehicle.

“A human expert, assuming 10 seconds per image, would take more than 15 years to perform the same task,” say Gebru and co.But how does the population of vehicles in an area relate to the local demographics? To find out, the team trained another deep-learning algorithm to learn the correlation between vehicle types and the data from U.S. Census and presidential election voting patterns in each precinct (an area of about 1,000 people).

Super Saurio Fly: Jurassic EditionSuper Saurio Fly is a prehistoric 2D arcade adventure game with a twist: Super Saurio is able to fly. Dodge traps and defeat other dinosaurs across sixteen different levels in a challenging journey around the world accompanied by hand-drawn animated cutscenes. Developer: Publisher:Genre:,Languages:,Tags: (31), (31), (31), (16), (16), (12), (12), (11), (11), (11), (11), (11), (10), (10), (10), (10), (10), (10)Category: Single-player, Multi-player, Co-op, Shared/Split Screen Co-op, Steam Achievements, Full controller support, Remote Play TogetherRelease date: Apr 12, 2018Price: $9.99Owners: 0. 20,000Followers: 0Peak concurrent players yesterday: 1. Super saurio fly jurassic edition movie online.

This training data set consisted of the data from 35 cities.They then used the rest of the data to test the deep-learning algorithm. The question they wanted to answer was: given the pattern of vehicles in an area, could the algorithm accurately predict the demographics as recorded in the U.S. Census and presidential voting data?It turns out that the deep-learning algorithm can do this remarkably well. “Using the classified motor vehicles in each neighborhood, we infer a wide range of demographic statistics, socioeconomic attributes, and political preferences of its residents,” they say.For example, sedans are most closely associated with Democrats, while extended cab pickups are more closely associated with precincts that voted Republican. “We found that by driving through a city for 15 minutes while counting sedans and pickup trucks, it is possible to reliably determine whether the city voted Democratic or Republican,” say Gebru and co.An important question is how well these conclusions match those gathered by surveys in the conventional way.

To find out, Gebru and co compared their results with those from the U.S. Census Bureau’s own data from the American Community Survey. And they found a strong correlation with demographic factors such as income, education, occupation, and so on. They were even able to make fine-grained predictions about voter preferences at the level of precincts, which encompass about 1,000 people.The American Community Survey is likely to remain the gold standard for data collection, at least for the time being.

But the ability to gather high-quality data quickly using Google Street View has great potential to provide almost real-time monitoring of changes in the population.And of course, Google is not the only organization collecting images of the streets—it’s not hard to imagine the same process being applied to images from Facebook, Twitter, or other social media networks that are geotagged and date-stamped.Ref:: Using Deep Learning and Google Street View to Estimate the Demographic Makeup of the US.