Welcome to the NNDB MAPPER!

The NNDB Mapper allows you to explore visually by graphing the connections between people and more.

After clicking the link above, select option A

Search for the Movie “Serenity” and then expand the nodes to connect it to the TV show “Firefly”

Another interesting way of searching visually with a map in order to discover relationships quickly.

Here are the current Top 10 Alt Search Engines. Cast your vote at TheSearchRace.com

A review by ReadWriteWeb’s Josh Catone.

According to a comScore study done last year, booking travel over the Internet has become something of a nightmare for people. It’s not that using any of the booking engines is difficult, it’s just that there is so much information out there that planning a vacation is overwhelming. According to the comScore study, the average online vacation plan comes together through 12 travel-related searches and visits to 22 different web sites over the course of 29 days. Semantic search startup UpTake (formerly Kango) aims to make that process easier.

UpTake is a vertical search engine that has assembled what it says is the largest database of US hotels and activities — over 400,000 of them — from more than 1,000 different travel sites. Using a top-down approach, UpTake looks at its database of over 20 million reviews, opinions, and descriptions of hotels and activities in the US and semantically extracts information about those destinations. You can think of it as Metacritic for the travel vertical, but rather than just arriving at an aggregate rating (which it does), UpTake also attempts to figure out some basic concepts about a hotel or activity based on what it learns from the information it reads. Things such as, is the hotel family friendly, would it be good for a romantic getaway, is it eco friendly, etc.

To read the rest of Josh’s review, click here.

AltSearchEngines obtained an exclusive interview with Elliot Ng, VP of Marketing for UpTake.com. Elliott has co-founded two successful ventures, Netcentives and Loyalty Matrix. Most recently Elliott ran web marketing for Intuit QuickBooks. Elliott started his career as the product manager for PowerPoint and Excel at Microsoft. Elliott is a graduate of Harvard and Harvard Business School.


Q: ASE: Elliott, how is UpTake’s vertical search the same or different from general search engines like Google?

A: Elliott Ng
As you know, all search engines do two things: matching, and ranking. Matching is the process of associating queries with the right documents. Ranking is ordering the documents based on their relevancy to a specific query term.

Statistic matching and ranking vs. semantic matching and ranking
General purpose search engines rely primarily on a keyword-based, statistical matching approach. There is no need for machine understanding of the concepts behind the keywords. UpTake.com uses a semantic matching and ranking approach. We believe this is a practical approach for a vertical domain (like travel) where its easier to use semantic analysis to understand the user intent behind queries, and also easier to extract meaning from documents.

Stateless search vs. conversational search sessions
The other big difference in our approach is that we can aim for a successful, multi-query, conversational search session. We can try to understand user intent as the user moves from general queries to more specific queries. Our vision of a conversational search experience is an achievable one because the context is someone’s search is known: they are planning travel. Even with the great advances Google has made in personalization and capturing Web history, the breadth of usage on Google for any given user means that Google must be very careful in trying to associate one query to another.

Q: ASE: Why do you think UpTake can deliver better results from a general search engine like Google?

A: Elliott Ng
The short answer is because we are focused on travel, so we can understand user intent better and analyze travel-related documents better.

Here’s the long answer: three interrelated factors help us deliver more relevant results and a better customer experience.

First, we’ve developed a domain-specific ontology. As you know, an ontology encompasses a set of concepts, relationships between concepts, and rules that can be applied to those concepts. The more specialized the field, the easier it is to create an ontology. Relative to the ontologies produced via academic research, what we are doing is very simple and practical. Our seed ontology is based on human expert knowledge. It has thousands of concepts, relationships and rules. Over time, our vision is to introduce machine-learning to automatically identify more relationships and propose more rules.

Second, we are conducting semantic analysis on a large number of documents. We only say the word “semantic” among search geeks and semantic Web fanboys like your audience! Here’s what we are doing:

Named-entity extraction. A simple example is we can identify a hotel or travel attraction and also extract all of that hotel or attractions’ attributes. This is made possible because we already understand how concepts are interrelated thanks to our travel ontology. We also identify travel reviews and opinions.

Sentiment analysis. We are using linguistic methods to analyze opinions and reviews to determine what a sentiment refers to (is it “kid-friendly”, “pet-friendly”, “dirty”, or “has a view”) and whether or not it is positive or negative.

Third, we believe that we can better understand user intent. How? In two ways: better query parsing and a conversational search approach.

Query parsing. Because we are talking about a very specific context, we can better look at query terms and understand them as concepts rather than just keywords. These concepts all come from the travel ontology so the relationship of these concepts to others are well known. For example, if someone types in “toddler-friendly” we know that this relates to “kid-friendly”, and that families generally appreciate “free-breakfast” and “suites.”

Conversational search approach. Vertical search engines have the potential of allowing users to express their intent in different ways. One way is to take a conversational search approach. The simplest way we are doing this is providing a rich set of refinement controls without typing in an entirely new search query as they would have to do in Google. Over time, we have an ambitious vision to allow users to highlight what they like and don’t like so we can provide more tailored recommendations to them. But we’ve only just begun on this part of the vision

Q: ASE: what are the biggest challenges for UpTake?

A: Elliott Ng
From a business standpoint, our biggest challenge as a brand new site is getting traffic. As we mentioned last week here in ASE, we firmly believe we live in a Google world and have optimized our search application to get traffic from natural search in Google. We do not think the host of new online travel sites are competitive. We don’t aim to be a travel online community or social network, and in fact think we can be very complementary to both new and old travel sites like these.

From a technology standpoint, we have plenty of work before we achieve our vision. We will continue to add more lodging types beyond hotels, and work to deliver the best search for things to do and attractions on the Web by the end of 2008! We also have some ideas around how we can help people decide “where to go”—no one on the Web truly delivers decision-support around picking your vacation destination.

Monetization is not a concern, even in difficult times like these. Travel lead generation is about an $8-10 billion market. And there is still about $300 billion of travel attractions and lodging that are not booked online, so there is still plenty of growth opportunity here.

Q: ASE: Well, since you work for a Travel Search engine, what’s your favorite travel destination?

A: Elliott Ng

By far the most memorable trip was to East Africa. My wife, Karen, and I traveled with a referred tour operator, Thomson Safaris, through some great game parks, including: Tarangire, Arusha, Serengeti, Kilimanjaro and Ngorogoro Crater. We also climbed Mt. Kilimanjaro; I made it to Barafu Camp at 15,600 feet (altitude pains) and Karen reached the summit at about 19,000 feet. After this, we lay on a white sandy beach in Zanzibar and recovered from what was the most difficult physical experience of my life!

By: Penny Herscher

A gauntlet was thrown down in the recent Techcrunch post, “Is Keyword Search About to Hit Its Breaking Point?” With a provocative up-and-to-the-right Web generations curve, the article claimed that the semantic web is the solution to the limitations of keyword search. And it’s coming in 2010 – if we’d just adopt the standards that would help computers extract meaning from the web.

I’m a believer in the general case, eventually. But, in many important domains, the solution is in production use today. Keyword search is a fantastic technology when a few words and popularity drive you to the right answer. It is a lousy solution when the answer involves an obscure concept or a relationship with common keyword characteristics. Unfortunately that is the case for most qualitative (text-based) business information. The semantic web is held up as a solution but, prior article aside, most in the field believe we’re a long way from this goal. Even is semantic tagging is enough, relying on people isn’t scalable or precise. Using machines is not yet reliable.

A real world solution
But there is a way to solve the problem. Decide on a domain and then solve within that domain.

This is being done effectively today for the investment and market research community today using what we call search-driven research. In the world of the professional investor or the executive, business decisions rely on fresh, conceptual information. Since it is both subtle and obscure, that information is extremely hard to find. Keyword search is not effective and a new paradigm is required. Truly valuable insights typically involve entities and their relationships - complex concepts that can’t be captured in keywords. Fortunately, they can be discovered once you’ve correctly identified and formalized the relevant concepts in each domain. The process demands detailed models which go way beyond keyword and tagging exercises.Investment and market research is a classic long tail problem; the high end of the Pareto curve contains the least empowering information. The search solution requires a shift from using prominence and popularity as a proxy for value to using business impact as a proxy instead.

For example – consider the portfolio manager who holds a pharmaceutical stock and wants to stay current on the market trends that affect his portfolio. Keyword search around, say, “drug discovery” or its synonyms produces no significant results beyond general education and company references. However, when conceptual knowledge about the domain is modeled, the search results can be high precision and value for the user. Actionable information discovery must happen at a much higher conceptual level.

Three step solution
The solution to model markets requires three technology families:

First - modeling the entities in a market. For business and investment decision makers the key entities – often colloquially called “topics” – include concepts like companies, market trends, management, brands or supply chain members – to name a few. The models of these entities drive detection based on a combination of keywords, grammar and natural language processing – and the web results are organized and tagged with the entities or business topics.

Second - modeling relationships between entities. In addition to modeling the entities, modeling the relationships between entities reveals the meaning. Relationships have many different types – for example status relationships such as competition between products or, in contrast, action relationships such as a merger between two companies or an executive moving from one company to another.

Third – detect events by identifying temporal patterns in the relationships between entities – detecting changes in how entities are arranged relative to each other in time.

Of course, all this wizardry needs to be combined with pragmatism to be useful in business – removing duplication, de-junking, factoring in source authority, normalizing and ranking results– to produce meaningful results from the unruly beast.

But the end result achieves the objective of the semantic web – to be able to use the web as a vast database of rapidly-changing, useful, decision making information. So rather than describe keyword search as reaching a breaking point, I’d describe it as useful for the problems it’s solving today, but not able to scale to the next class of business problems which can be solved using a domain-specific approach today.