Features

Jennifer Zaino
SemanticWeb.com Contributor

Newcomer semantic social search engine TipTopSearch—the brainchild of TipTop Technologies and ITC Infotech—wants to help make your Thanksgiving festivities just that much more special, with the help of the Twitter community. Head over here for its Thanksgiving search special. It uses its real-time sentiment, opinion and experience analytics, attribute sorting and semantic filtering functions, and ranking and relevancy algorithms to dish up ideas about what should be on the menu for the holiday to help hosts and hostesses whip their tables into shape.

The everyday goal of the recently launched service is to contextually deliver the insight and experiences of others that will help make people make decisions in their personal lives by taking advantage of the unstructured data that lives out there on the social web, as well as share that with their own networks.

For example, type in "Fantastic Mr. Fox" – what many parents will be doing with their home-from-school kids the day after Turkey Day—and you’ll get a set of results from current conversations underway on the subject at Twitter. Those results are based on the engine’s search of messages there, its “reads” of what is being said, and then its extraction of what it considers to be “the most interesting, useful and unique messages that match the search term,” according to its FAQ.

continued...

Jennifer Zaino
SemanticWeb.com Contributor

Swine flu has been sweeping the nation – or at least the news media. Some of the most recent reports put 22 million Americans as having suffered through H1N1, with 3,900 of them having died as a result of swine-flu related causes.

With flu season already underway—and with peak flu season just around the corner in January—Semanticweb.com thought we’d explore how some of the new health sites that leverage semantic web concepts are helping their users manage through the tough times. Below we present three.

HealthMash

The interface provides a very clean look with easily understood tabbed access to the multiple ways you might want to find information about the bug: Trusted health information, recent news, video, images, and blogs – not to mention drugs and substances, clinical trials and integrative medicine for the alternative slant. One of the core principles of the service was to use quality information sources in combination with its semantic health knowledge base, and it appears to be more or less keeping to that idea of mashing information from reputable sites. Its blog recommendations may seem a bit limited for exploring the swine flu, but at least those in search of serious health information won’t be directed to the ho-ho-ho one about Santa facing swine flu concerns.

continued...

Jennifer Zaino
SemanticWeb.com Contributor

In Italy, Quattroruote is a leading online magazine for car aficionados and buyers, with its reputation built on testing and evaluating models and its own blue book-like price estimates for vehicles. Now it’s a leading-edge user of semantic web technology, too.

It has deployed Expert System’s Cogito semantic solution to help add value to user searches for used cars in its portal to the world of classified car sales.

“Semantic tools are extremely beneficial to being able to properly understand classified ads and categorize them correctly and accurately,” says Luigi Conti, director of publisher Editoriale Domus. “Semantic capabilities ease the way people interact with our search engine, and we want the search experience to be as easy as asking a question to a live person.”

The company has a marketplace where it offers traditional search services, but its semantically powered CheAuto “is a completely different story,” he says. CheAuto’s mission, Conti says, is to provide users a unique point of search for used car classified ads, adding two elements to the search results.

First, the amount the seller is looking to spend is compared with what Quattroruote deems to be the official value of the used car, when possible.” In Italy, the Quattroruote Used Car Value is a quasi-standard de facto,” he notes.

Second, all of the different classified ads are categorized and ordered according to Quattroruote Infocar, a database of all the cars (make, model, and so on) sold in Italy since 1980.

“CheAuto is a meta search engine that allows users to search not only within one marketplace, but within many of them, ultimately having insight into every significant marketplace in Italy,” Conti says.

continued...

Jennifer Zaino
SemanticWeb.com Contributor

How’s your RDF? If it could be in better shape, some folks may be able to help: The Pedantic Web Group was recently formed by researchers at Digital Enterprise Research Institute (DERI) and Institute AIFB at the Universitaet Karlsruhe (see previous article).

To find out more about the new initiative, SemanticWeb.com recently conducted an email conversation with one of the group’s founders, Aidan Hogan, who is working on the URQ research stream at DERI, a work program to find the right trade-off between expressive knowledge representation and efficient, scalable reasoning and querying techniques in the open, distributed environment of the Semantic Web.

SemanticWeb.com: Have you and your colleagues observed a growing trend of RDF data being published? What might that lead you to conclude about the growing maturity of the semantic web?

Hogan: There is certainly an encouraging trend of growth in RDF data -- both in terms of quality, heterogeneity and quantity -- being published to the Web. Back in 2005 when I started working in the area of Semantic Web research, RDF Web data consisted of a number of interlinked FOAF profiles and some data published under the auspices of various research projects or geek curiosity. The quality of the data was, as I remember, quite poor. Publishers were reluctant to use URIs to name their resources, vocabularies were replete with errors, interlinking between datasets was either poor or nonexistent. My own FOAF file, at that time, was no different; they were certainly more innocent times.

Jump to late 2009 and we've come a long way. More specifically, under the pragmatic guidance of the Linked Data movement, RDF data published on the Web has come a long way. The Linked Data movement has been integral to the maturation of RDF Web publishing, not merely by promoting a set of pragmatic best practices, but also by refocusing efforts on producing data: before, data was often published in RDF as an afterthought, or for the purposes of a specific application. Now, Linked Data advocates publishing RDF data on the Web as a worthwhile endeavor in itself. As such, we are now on our way to solving the chicken-and-egg problem with the Semantic Web with respect to which comes first: the data or the applications. 

Now is an exciting time for R&D into applications which can exploit the fruits of Linked Data. As compared to four or five years ago, data quality has improved as, for example, publishers understand the importance of using URIs to name things, and that those URIs should be dereferencable. Quantity and heterogeneity has also increased, as the March '09 LOD cloud [refer ttp://linkeddata.org] can attest to; data is being published by governmental and commercial entities and is becoming more 'general-interest'. And, the trend is continuing; for example, there were two exciting announcements at ISWC last week: that Drupal 7 core will support SIOC/RDFa exports by default and that the New York Times are planning to produce Linked Data exports.

continued...

Jennifer Zaino
SemanticWeb.com Contributor

The Book of Odds web site offers a forum for determining the odds of everyday life (see previous article), letting users combine odds statements in unexpected ways in real time to compare, for example, whether there’s a greater risk of dying from an encounter with a shark or a vending machine (it’s the latter).

It’s interesting that the site is powered by Cambridge Semantics technology, because that vendor sees its business customers increasingly eager to take advantage of its tools for combining, using and sharing data from disparate sources - regardless of variations in data structure - to deal with their business risks.

Along with growing their toplines and controlling costs, risk is one of the most prominent issues Cambridge customers want to address, says CEO Michael Cataldo.

“And the biggest risk is associated with questions you just don’t know and so you can’t answer,” he says. “One of the reasons I think semantics is going to be so hot is that it puts the capability of answering those kinds of questions in the hands of the end users.”

continued...

Jennifer Zaino
SemanticWeb.com Contributor

At last week’s International Semantic Web Conference (ISWC 2009), semantic web application development vendor TopQuadrant announced oeGOV. The initiative aims to create an open, W3C Semantic Web standards-based set of ontology models to encourage and facilitate the use of linked government data.

SemanticWeb.com conducted an email conversation with TopQuadrant’s Dean Allemang, chief scientist, and Ralph Hodgson, co-founder and CTO, to learn more about the effort to help the government sector embrace semantically linked open data.

Semanticweb.com: Why does TopQuadrant see a need for this initiative?

TopQuadrant: Open Linked Data in the government is taking off in a big way. But there is a lot of data out there that has to be presented in a coherent, reusable way. Ontologies help that happen by providing support for aggregation, provenance and data quality – aggregation through everything having URIs and controlled vocabularies, provenance in terms of who was the source and when did the data appear, quality through units of measure and data types. Government data is available to the public, but not easily accessed at the moment. It isn’t even very big data – there aren’t any serious technical barriers to making it more available. We feel that now is the time to start up an initiative to bring public data to the public.

continued...

Jennifer Zaino
SemanticWeb.com Contributor

Slowing down their start-up investments, VCs aren’t raising as much money these days. Recently Thomson Reuters and the National Venture Capital Association said the third quarter saw the smallest number of venture funds raising money in a single quarter since the third quarter of 1994.

Just 17 venture capital funds raised $1.6 billion in the third quarter this year – in 1994’s third quarter, 17 funds were also raised and the lowest level of dollars were committed since the first quarter of 2003 when $938 million was raised during the dot-com bust. The NVCA says it expects commitment levels to remain modest the rest of the year, with gradual increases beginning in 2010.

Given the state of VC fund-raising and funding, semantic web start-ups perhaps won’t be as well-served by traditional West Coast venture capital firms as they may have been a couple of years ago. So why not try Houston instead?

That’s the location of Creeris Ventures, which is backing some innovative start-ups, including 80legs. In fact, if you happen to be a semantic web business that could leverage 80legs’ scalable web content crawling and processing service, you may want to give Creeris CEO Brad Wilson a call.

“We don’t have any semantic web technologies or companies in our portfolio now,” he says, “but that’s not to say we wouldn’t be interested in either forming or partnering with people [in this area] to create new companies...especially those that would make sense in combination with 80legs' core technology to drive 80legs usage and possibly higher margins.”

Also note he’s not interested in scientific type-research projects, which may be cool but have limited applicability – only those businesses that have potential for in-demand commercial applicability need apply.

(Speaking of 80legs, that company has just launched a developers’ challenge to create applications that use the platform they can sell — and keep 100 percent of the profits for — for its App Store that opens in November. The idea, 80legs says, is to make crawling even more accessible — so as to expand the market to the non-technically inclined — with a rich store of 80apps, so that anyone will be able to execute jobs.)

continued...

Jennifer Zaino
SemanticWeb.com Contributor

You’ve heard of the Book of Love – now there’s also a Book of Odds.

Don’t take out your betting cards yet. Rather than being a tool to help you get lucky on the next horse race, Book of Odds is about providing the odds of everyday life. It defines itself as an ever-expanding semantic database of probabilities that helps users find unexpected links between odds statements – the odds of a man over 50 being diagnosed with heart disease is about the same as the odds that an adult is afraid of snakes (1 in 1.9x).

Launched in mid-October, the site claims to contain hundreds of thousands of carefully researched Odds Statements, each evaluated and graded for its underlying data quality. The site uses Cambridge Semantics’ semantic middleware technology to structure the data to help users in their searches for odds statement matches. Currently the site has four topic portals – Health & Illness, Accidents & Death, Relationships & Society, and Daily Life & Activities, each with their own sub-topics.

Users can search by keywords, key-numbers (finding events of a given likelihood through whole number searches), and visually. Semanticweb.com thought this presented lots of possibilities for whiling away a morning, so we thought we’d share with you some of the interesting odds we came across while perusing the site. Check them out on the jump, right below the screen shot of the Book of Odds home page.

continued...

XMLToday.org managing editor Kurt Cagle reports that he gave a presentation earlier this month at the Online News Association conference in San Francisco on how the semantic web will impact journalism:

The talk was only semi-successful - due to a number of factors (not least of which being a parade of revelers passing in front of the conference hotel wearing little more than paint ... and sometimes not even that) I had a number of people abandoning the talk about halfway through. However, I think the talk provides a good overview of how Semantic Web technologies factor into the field of Journalism ... though without the nude party-goers.

It's hard to believe that Drupal references and taxonomy definitions can't compete with naked, but based on Cagle's account, such appears to be the case.

Cagle's slide presentation is embedded below. It's very thorough and interesting, especially his view of the evolving role of journalists ("The significant journalists today are analysts. Their role today is to discern meaning and validity in a rushing tide of assertions. Increasingly less important is their role as reporters.") and how semantic technology is enabling this transition.

Oh, in case you were hoping, Cagle was good to his word: There are no nude party-goers in the slide show. You'd think he would have learned something from his recent experience.

Jennifer Zaino
SemanticWeb.com Contributor

The World Wide Web Consortium (W3C) in late September chartered the Provenance Incubator Group, charged with developing a road map for Semantic Web technologies. Chairing the group is Yolanda Gil, associate director for Research, Intelligent Systems Division, and research associate professor of computer science at the Information Sciences Institute at the University of Southern California.

The importance of delving into the issues behind understanding and verifying the origin of information speaks to one thing: The semantic web has arrived.

"Provenance is a very practical matter, and if we didn't have large data available on the semantic web and people interested in using it and reasoning with it -- if we didn't have scientists who have adopted semantic web technologies and want to solve a big problem they have for integrative science -- if we didn't have such a push, I don't think we would be thinking about this so hard or starting this effort," Gil says. "I think it attests to just the way the semantic web is spreading and getting adopted and finding its way to real use."

Information provenance, she explains, is an issue for everyone, whether it's a researcher sourcing data on the web, a consumer searching social webs for information on vacation destinations, or scientists trying to connect their data findings with the work of their peers.

"We all do analysis, but there's nothing on the web that represents that," she says, when it comes to those ways of determining whether we trust certain information. Were there a way to access exactly what the provenance of information is, such as who was the original source of certain news, for the purposes of more accurately assessing that data, it would have wide-ranging benefits. "Browsers could potentially exploit that but also semantic reasoners could use it," she says, to help users determine which is the valid source when there are discrepancies around information that is presented as fact but is inconsistent among different web sources.

The growth of linked data adds to the provenance challenge.

"Now that there's all this data linked together, you start asking questions and then you have to take into account where the information came from, and that it's not just a level field for every single piece of data in the network," she says. And provenance really gets interesting as it relates to the scientific community, in areas such as life sciences. (Gil's research activities in this area have resulted in papers including Examining the Challenges of Scientific Workflows.) As an example, she points to work around genomics involving population studies across a broad range of populations to determine linkages among genes and certain diseases. In such work, genotypes are likely collected at different sites, perhaps using different machines or even different settings. "Unless you have proof of how genotype information came to be and how it was created, you could mix apples and oranges in your analysis," she says, potentially resulting in inaccurate leads.

"In science every day there is a bigger push to integrative science, to integrate information from diverse sources," she adds, which increases the requirement to ensure scientists are integrating meaningful data for their computations. Scientists also are very concerned about reproducibility, a cornerstone of the scientific method that provenance recording can help preserve, she says.

"We have a long history of looking at what are the human abilities that are useful to us [in assessing trust in information], and can we try to come up with reasoning mechanisms we could use on the machines, on the semantic web," she says. "It's very early on, but imagine eventually having some kind of standard language to express that this web site has original content that I created, this is my signature, you can prove that it is me stating this content, and if you find this content reproduced elsewhere you know I created it and it's authentic."

You'd also know that Gil is a researcher at the University of Southern California, adding to her weight as a trustworthy source for information relating to computer science and AI, for instance. "You can use that information to make judgments," Gil says.

Mechanisms for determining provenance also could have applicability for safeguarding intellectual property and copyrights online. "As people copy images from the web, provenance again might be useful here to figure out, with that information, that there is a restriction about using an image for a particular purpose," she says. "A system might automatically warn you of those kinds of things."

The incubator will get started by posting use case scenarios to illustrate the range of uses for provenance information, and then figure out from there what semantic web representation requirements might be needed to address those scenarios. As part of that the group may liaison with other W3C groups to address related issues that might be better addressed by other parts of the web architecture. Take the example of two web sites, each providing different values for the height of the Empire State Building. The work of the provenance group could lead to mechanisms for some day identifying that one of the sites is repurposing data originating on a high school kid's class project web site, while the other repurposes information from the Visitors' Bureau of New York City. But the W3C Security Activity group may be the more appropriate party to police whether that a site claiming to be Visitors Bureau really is the Visitors' Bureau.

So far the Provenance group has about 15 members with more joining.

"The important thing," says Gil, "is we have representation from a broad variety of areas, from database professions that have been looking at issues of provenance, from the argumentation tradition of how you represent and reconcile different points of view, from people in life sciences research like myself bringing their perspectives. To me that's what's really exciting."