(Reuters) - Tired of digging through long-winded restaurant reviews to find a great meal? Next time, turn to your smartphone where personalized search engines will lead your stomach in the right direction.
A newcomer in the world of personalized search, Ness Computing recently released a free iOS app that provides restaurant recommendations based on a user's personal tastes and information from friends gathered through social media sources such as Foursquare and Facebook.
The app uses that information to calculate a "Likeness Score," which is the probability a user will enjoy a particular venue. Integration with social networks is optional, however, it is one way Ness can achieve more relevant results.
"There's all sorts of information that people's friends have left when they check in (to a venue), or mention 'I'm having a great meal at this place'", said Corey Reese, co-founder and CEO of Ness Computing. "We wanted to build a beautiful interface for people to find that kind of content and information on their mobile device."
The app, which some bloggers compare to Netflix (movies and TV) or Pandora (music) for restaurants, provides information such as addresses and phone numbers, check-ins, comments and tips from social networks. Check-ins amongst friends influence a restaurant's ranking in search results.
The app provides the ability to filter out major chains, which appeals to people looking for independent restaurants. And to facilitate finding new places to eat, users can hide restaurants they have already rated. Search results can also be filtered by distance and price.
Reese said user design has been a core focus for Ness. Looking to Apple as a model, the company hired Scott Goodson as Director of Mobile Engineering, a former engineer at Apple who was one of the first members of the iOS team and helped to build some of its original flagship apps.
"When you talk to the folks at Apple on their design and engineering teams, it's almost a religious experience to make this stuff. And we think part of the reason they've been so successful is because they take their product development so seriously", said Reese.
The underlying technology took 18 months to develop with a lot of effort put into refining an index of restaurants and standardizing data from sources like Foursquare, Facebook and CitySearch.
"When we ran one of our first experiments, McDonald's was obviously one of the most popular restaurants to show up in our system. There were 1,900 ways that our data sources had spelled McDonalds," said Reese. "We had to clean all of that data so that there was only one way of spelling McDonald's and then associate all that data with each other."
There are many personalized search engines and apps available. Microsoft's Bing, which markets itself as a "decision engine", also taps into Facebook data when generating search results. Mobile app Alfred and website Hunch also hook into social networks and provide similar functions.
According to Reese, upcoming plans for Ness include sentiment analysis of data from social sources, which is the ability to automatically determine whether a particular comment is positive, negative or neutral. They also plan to increase Facebook support and implement Twitter integration.
The company plans to expand to other vertical markets, such as shopping, by the end of the year before expanding overseas.
The app is currently available on the iOS platform on the Apple store, and only in the US.
A newcomer in the world of personalized search, Ness Computing recently released a free iOS app that provides restaurant recommendations based on a user's personal tastes and information from friends gathered through social media sources such as Foursquare and Facebook.
The app uses that information to calculate a "Likeness Score," which is the probability a user will enjoy a particular venue. Integration with social networks is optional, however, it is one way Ness can achieve more relevant results.
"There's all sorts of information that people's friends have left when they check in (to a venue), or mention 'I'm having a great meal at this place'", said Corey Reese, co-founder and CEO of Ness Computing. "We wanted to build a beautiful interface for people to find that kind of content and information on their mobile device."
The app, which some bloggers compare to Netflix (movies and TV) or Pandora (music) for restaurants, provides information such as addresses and phone numbers, check-ins, comments and tips from social networks. Check-ins amongst friends influence a restaurant's ranking in search results.
The app provides the ability to filter out major chains, which appeals to people looking for independent restaurants. And to facilitate finding new places to eat, users can hide restaurants they have already rated. Search results can also be filtered by distance and price.
Reese said user design has been a core focus for Ness. Looking to Apple as a model, the company hired Scott Goodson as Director of Mobile Engineering, a former engineer at Apple who was one of the first members of the iOS team and helped to build some of its original flagship apps.
"When you talk to the folks at Apple on their design and engineering teams, it's almost a religious experience to make this stuff. And we think part of the reason they've been so successful is because they take their product development so seriously", said Reese.
The underlying technology took 18 months to develop with a lot of effort put into refining an index of restaurants and standardizing data from sources like Foursquare, Facebook and CitySearch.
"When we ran one of our first experiments, McDonald's was obviously one of the most popular restaurants to show up in our system. There were 1,900 ways that our data sources had spelled McDonalds," said Reese. "We had to clean all of that data so that there was only one way of spelling McDonald's and then associate all that data with each other."
There are many personalized search engines and apps available. Microsoft's Bing, which markets itself as a "decision engine", also taps into Facebook data when generating search results. Mobile app Alfred and website Hunch also hook into social networks and provide similar functions.
According to Reese, upcoming plans for Ness include sentiment analysis of data from social sources, which is the ability to automatically determine whether a particular comment is positive, negative or neutral. They also plan to increase Facebook support and implement Twitter integration.
The company plans to expand to other vertical markets, such as shopping, by the end of the year before expanding overseas.
The app is currently available on the iOS platform on the Apple store, and only in the US.
