In-Brief: The collaboration between AltSearchEngines and MSG gets into gear with a look at BooRah, a restaurant search engine that aggregates restaurant review content and buzz to let users search for what they want how they want it. The approach covers the bases to revolutionize how we might find businesses and services on the fly. Thanks to Nagaraju Bandaru, BooRah Co-founder & CTO, for an excellent webinar, as well as an exciting preview of what’s in the pipeline.

At first glance, BooRah may appear to be a straight-forward restaurant search engine. But take a look under the hood and you’ll find a patent-pending system at work that effectively analyzes user-generated content (restaurant reviews, blogs etc…) to pick up on positive and negative sentiment. Hence the name: Boo refers to negative opinion (what people don’t like), and Rah tells us people gave it a thumbs-up – as in Hoorah!

Equipped with this information – which the technology gleans from semantic and structural analysis it performs on every sentence to identify the adjectives, verbs, and nouns that tell us what others really feel – the search engine lets users find restaurants they’re sure to like (based on factors including food, service, and ambience).

Pair this with mobile, as BooRah did just a few weeks ago, and you have a smart service that is location-aware and buzz-sensitive.

The service is in Alpha and will need a few more features before the company can call it a full Beta. Nagaraju tells me he just added directions and will continue to add features. “Basically, anything you can search for at BooRah.com, you can search for using mobile search.”

MOBILE DECISION-MAKING TOOLS: BooRah taps the wisdom of crowds to bubble up relevant answers and recommendations. “When you are looking for a restaurant, you want to do more than eat. You want to know if it’s a hip place where you can gather with friends for a casual meal, or an upscale restaurant that would be a perfect place to take your date. You want to know that without reading lots of reviews. And on mobile, the relevance that we can bring to results is even more critical.” To avoid information overload, BooRah presents 10 mobile search results. The presentation is neat and clean, and leaves room for related advertising – when the time is right.

NUTS & BOLTS: By way of background, BooRah’s semantic smart search system is built from the ground up to let users perform natural language search for specific experiences, moods and sentiments, and personalize the results. A query for “anniversary in Palo Alto, CA” would search for all relevant comments associated with categories such as “romantic”, “upscale”, “trendy”, “Live-music”, “view”, etc… and lets users filter results based on their preferences.

BooRah’s Natural Language Processing (NLP) technology processes vast amounts of plain English text from user-generated content such as user reviews, opinions, and blogs. It extracts specific user sentiments expressed within the content and generates scores, summaries, and customizable search responses relevant to the consumers. Results are ranked first and foremost according to popularity, and BooRah purposely ignores stars and similar vague and subjective forms of user ratings. A wise move since people rarely give out one star or five – it’s always somewhere in the middle. Rather than wade through this – and risk skewing results in favor of hyperactive contributors or users with ulterior motives for giving thumbs-up or thumbs-down – BooRah hones on opinionated reviews that say why a restaurant is good or bad.

LOCATION: BooRah on mobile handles several different types of queries, including street name, full address, zip code, points of interest. One I wasn’t expecting is searching by cross streets and intersections. “This is how we believe people will search. Let’s say you are standing at the intersection of El Camino and Castro Street, then you will want to know what is nearby … and we can present results according to distance as well as user review sentiment.” BooRah will also leverage GPS on mobile devices and “provide an option where people who are integrating with our content will be able to leverage that (GPS) as well.”

Ironically, BooRah started out in mobile and with a strong emphasis on local search, so expect more features that go far beyond plain-vanilla (what’s nearby) local search. Functionality high on Nagaraju’s agenda is mobile social networking and a way to show and share searches as part of the user profile.

LONG TAIL SEARCH: During the demo, this feature stood out from the rest. Long tail search meets local to tell me where I can get the XXX (fill in the blank). In our demo, we searched for “foie gras”. The results only included restaurants where users had had a positive experience related to foie gras. To be clear – this wasn’t a search that turned up where a user referred to foie gras; BooRah’s technology exposed results where users specifically liked the food. It all goes back to BooRah’s patented summarization algorithm that combines statistical NLP techniques with knowledge-based methods of processing simple text – and it’s a far cry from what Google & Co can deliver.

SYNDICATION: BooRah mobile is a destination, but the company has aspirations that go beyond becoming a household name for restaurant search. Nagaraju tells me the best strategy may be one where BooRah is never mentioned. “We can power the companies [search engines, content providers, etc...] with the content [reviews]. We don’t want to be the one that has to get the experience right on the mobile phone, because there are a lot of people who can do that better. We want to be the content provider for those companies, the company they partner with, because then the pie is bigger for us both.”

This model is so new that Nagaraju had to come up with a name for it. He chose “Mobile Search Syndication Provider.” As he sees it: The model is quite similar to the one already in place between BooRah and newspapers online. “The approach carries over to the mobile phone…. They [partners] get the content and we do a rev-share.” BooRah is already in discussions with search engine providers, as well as one start-up company, working on an iPhone app. Nagaraju declined to give more detail, but he did note that BooRah is “definitely leaning toward integration with Google, especially on the iPhone.” When? No set timeframe – but we can look for maps and a few other cool features in June.

MONETIZATION: It’s early days, so Nagaraju is looking for some proof points before thinking about how to make money through search advertising. “Local is a gold mine if you can figure out what the product is, because these companies (restaurants, local businesses, etc…) are willing to pay.” To date, BooRah is in discussions with directory providers and mobile operators, but nothing to report just yet…

How would BooRah monetize its search service in practice? We mused about several approaches, including click-to-call and other mechanisms that would yield likely high CPMs and pave the way toward a pay-per-transaction model. But it doesn’t stop there. “We have discounts for about 20,000 restaurants. And people are actually looking for restaurants, so we can put a discount certificate to a user when they are in the proximity and say, hey, why don’t you walk into this restaurant and then you can get a restaurant.com discount certificate; a twenty-five dollar certificate for ten bucks? That’s highly relevant to the user experience because they’re actually getting something that they’re looking for. We get a cut off of 10-20 percent of that, the restaurant gets the customer, and the affiliate has made money as well.”

SOCIAL STUFF: Nagaraju has some cool ideas about community and gave me a preview of where his thinking could soon lead BooRah. He showed me a prototype service where he has listed himself as a fan of a particular restaurant. “First, we can abstract that data and very easily provide that as a recommendation.” Take it a step further and the restaurants I search for – and the results I get – are part of my profile that I can share in a new kind of social recommendation.

A light goes click in my head. We have mobile search with more than a community-feel; we have a feedback loop and no need to always start from square one with every new search. We can refine our search results. How does this work? Let’s say Nagaraju goes out and does the heavy-lifting, searching until he gets the best list of places where I can enjoy classic Lebanese cuisine. The search results are spot-on and now I don’t have to duplicate the search. I can use his results and use that as a starting point to look for more of the same – or explore what else he likes since I feel we have similar tastes. Nagaraju tells me BooRah is several months away from harnessing some sort of social search and recommendation. But it is the direction the company will go. Smart social search – a combination sure to yield relevant results every time…

EXPANSION: Nagaraju sees a “horizontal opportunity to take the engine to other categories such as travel and hotels.” It’s also possible to reach out beyond the U.S. to other English-speaking countries. (BooRah currently delivers information gathered from over 1 million online restaurant reviews in the top 20 metros across the U.S. including San Francisco, Los Angeles, and New York, and will expand to cover all of the U.S. in “the next month or two.”) Nagaraju says there is keen interest from Asia – particularly Korea – he’s also exploring.

Thanks to the many “Alts” (alternative search engine companies) that have contacted me to set up briefings. (My PA Andrea Henninge – andrea@msearchgroove.com is your point of contact.) The line-up over the next weeks comprises cool companies including Ziva, Hiogi, and Wispoon, as well as a fresh look at established mobile search engines such as Taptu, Medio, and JumpTap. I encourage you to check back regularly!

This entry was posted on Tuesday, June 10th, 2008 at 1:33 pm and is filed under Mobile Advertising, Mobile Search, Mobile Social Networking, Personalization, Recommendation. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.