One of the highlights of October's Web 2.0 Summit in San Francisco was the emergence of 'Semantic Apps' as a force. Note that we're not necessarily talking about the Semantic Web, which is the Tim Berners-Lee W3C led initiative that touts technologies like RDF, OWL and other standards for metadata. Semantic Apps may use those technologies, but not necessarily. This was a point made by the founder of one of the Semantic Apps listed below, Danny Hillis of Freebase (who is as much a tech legend as Berners-Lee).
The purpose of this post is to highlight 10 Semantic Apps. We're not touting this as a 'Top 10', because there is no way to rank these apps at this point - many are still non-public apps, e.g. in private beta. It reflects the nascent status of this sector, even though people like Hillis and Spivack have been working on their apps for years now.
Firstly let's define "Semantic App". A key element is that the apps below all try to determine the meaning of text and other data, and then create connections for users. Another of the founders mentioned below, Nova Spivack of Twine, noted at the Summit that data portability and connectibility are keys to these new semantic apps - i.e. using the Web as platform.
In September Alex Iskold wrote a great primer on this topic, called Top-Down: A New Approach to the Semantic Web. In that post, Alex Iskold explained that there are two main approaches to Semantic Apps:
1) Bottom Up - involves embedding
semantical annotations (meta-data) right into the data.
2) Top down -
relies on analyzing existing information; the ultimate
top-down solution would be a fully blown natural language processor, which is able to
understand text like people do.
Now that we know what Semantic Apps are, let's take a look at some of the current leading (or promising) products...
Freebase aims to "open up the silos of data and the connections between them", according to founder Danny Hillis at the Web 2.0 Summit. Freebase is a database that has all kinds of data in it and an API. Because it's an open database, anyone can enter new data in Freebase. An example page in the Freebase db looks pretty similar to a Wikipedia page. When you enter new data, the app can make suggestions about content. The topics in Freebase are organized by type, and you can connect pages with links, semantic tagging. So in summary, Freebase is all about shared data and what you can do with it.
Powerset (see our coverage here and here) is a natural language search engine. The system relies on semantic technologies that have only become available in the last few years. It can make "semantic connections", which helps make the semantic database. The idea is that meaning and knowledge gets extracted automatically from Powerset. The product isn't yet public, but it has been riding a wave of publicity over 2007.
Twine claims to be the first mainstream Semantic Web app, although it is still in private beta. See our in-depth review. Twine automatically learns about you and your interests as you populate it with content - a "Semantic Graph". When you put in new data, Twine picks out and tags certain content with semantic tags - e.g. the name of a person. An important point is that Twine creates new semantic and rich data. But it's not all user-generated. They've also done machine learning against Wikipedia to 'learn' about new concepts. And they will eventually tie into services like Freebase. At the Web 2.0 Summit, founder Nova Spivack compared Twine to Google, saying it is a "bottom-up, user generated crawl of the Web".
AdaptiveBlue are makers of the Firefox plugin, BlueOrganizer. They also recently launched a new version of their SmartLinks product, which allows web site publishers to add semantically charged links to their site. SmartLinks are browser 'in-page overlays' (similar to popups) that add additional contextual information to certain types of links, including links to books, movies, music, stocks, and wine. AdaptiveBlue supports a large list of top web sites, automatically recognizing and augmenting links to those properties.
SmartLinks works by understanding specific types of information (in this case links) and wrapping them with additional data. SmartLinks takes unstructured information and turns it into structured information by understanding a basic item on the web and adding semantics to it.
[Disclosure: AdaptiveBlue founder and CEO Alex Iskold is a regular RWW writer]
Hakia is one of the more promising Alt Search Engines
around, with a focus on natural
language processing methods to try and deliver 'meaningful' search results. Hakia attempts to
analyze the concept of a search query, in particular by doing sentence
analysis. Most other major search engines, including Google, analyze keywords.
The company told us in a March interview that the future of search engines will go beyond keyword
analysis - search engines will talk back to you and in effect become your search
assistant. One point worth noting here is that, currently, Hakia has limited post-editing/human interaction for the editing of hakia Galleries, but the rest of the engine is 100% computer powered.
Hakia has two main technologies:
1) QDEX Infrastructure (which stands for Query Detection and Extraction) - this does the heavy lifting of analyzing search queries at a sentence level.
2) SemanticRank Algorithm - this is essentially the science they use, made up of ontological semantics that relate concepts to each other.
Talis is a 40-year old UK software company which has created a semantic web application platform. They are a bit different from the other 9 companies profiled here, as Talis has released a platform and not a single product. The Talis platform is kind of a mix between Web 2.0 and the Semantic Web, in that it enables developers to create apps that allow for sharing, remixing and re-using data. Talis believes that Open Data is a crucial component of the Web, yet there is also a need to license data in order to ensure its openness. Talis has developed its own content license, called the Talis Community License, and recently they funded some legal work around the Open Data Commons License.
According to Dr Paul Miller, Technology Evangelist at Talis, the company's platform emphasizes "the importance of context, role, intention and attention in meaningfully tracking behaviour across the web." To find out more about Talis, check out their regular podcasts - the most recent one features Kaila Colbin (an occassional AltSearchEngines correspondent) and Branton Kenton-Dau of VortexDNA.
UPDATE: Marshall Kirkpatrick published an interview with Dr Miller the day after this post. Check it out here.
Venture funded UK semantic search engine TrueKnowledge unveiled a demo of its private beta earlier this month. It reminded Marshall Kirkpatrick of the still-unlaunched Powerset, but it's also reminiscent of the very real Ask.com "smart answers". TrueKnowledge combines natural language analysis, an internal knowledge base and external databases to offer immediate answers to various questions. Instead of just pointing you to web pages where the search engine believes it can find your answer, it will offer you an explicit answer and explain the reasoning patch by which that answer was arrived at. There's also an interesting looking API at the center of the product. "Direct answers to humans and machine questions" is the company's tagline.
Founder William Tunstall-Pedoe said he's been working on the software for the past 10 years, really putting time into it since coming into initial funding in early 2005.
Tripit is an app that manages your travel planning. Emre Sokullu reviewed it when it presented at TechCrunch40 in September. With TripIt, you forward incoming bookings to plans@tripit.com and the system manages the rest.
Their patent pending "itinerator" technology is a baby step in the semantic web - it
extracts useful infomation from these mails and makes a well structured and
organized presentation of your travel plan. It pulls out information from Wikipedia for the places that you visit. It uses
microformats - the iCal format, which is well integrated into GCalendar and
other calendar software.
The company claimed at TC40 that "instead of dealing with 20 pages of planning, you just print out 3 pages and everything is done for you". Their future plans include a recommendation engine which will tell you where to go and who to meet.
ClearForest is one of the companies in the top-down camp. We profiled the product in December '06 and at that point ClearForest was applying its core natural language processing technology to facilitate next generation semantic applications. In April 2007 the company was acquired by Reuters. The company has both a Web Service and a Firefox extension that leverages an API to deliver the end-user application.
The Firefox extension is called Gnosis and it enables you to "identify the people, companies, organizations, geographies and products on the page you are viewing." With one click from the menu, a webpage you view via Gnosis is filled with various types of annotations. For example it recognizes Companies, Countries, Industry Terms, Organizations, People, Products and Technologies. Each word that Gnosis recognizes, gets colored according to the category.
Also, ClearForest's Semantic Web Service offers a SOAP interface for analyzing text, documents and web pages.
Spock is a people search engine that got a lot of buzz when it launched. Alex Iskold went so far as to call it "one of the best
vertical semantic search engines built so far." According to Alex there are four things that makes their
approach special:
As a vertical engine, Spock knows important attributes that people have: name, gender, age, occupation and location just to name a few. Perhaps the most interesting aspect of Spock is its usage of tags - all frequent phrases that Spock extracts via its crawler become tags; and also users can add tags. So Spock leverages a combination of automated tags and people power for tagging.
What have we missed? ;-) Please use the comments to list other Semantic Apps you know of. It's an exciting sector right now, because Semantic Web and Web 2.0 technologies alike are being used to create new semantic applications. One gets the feeling we're only at the beginning of this trend.
Listed below are links to blogs that reference this entry: 10 Semantic Apps to Watch.
TrackBack URL for this entry: http://www.readwriteweb.com/cgi-bin/mt/mt-tb.cgi/1796
Wondering who the top applications are in the emerging semantic web field? The excellent Read/WriteWeb blog published a top 10 list last week. Freebase, Powerset, Twine, AdaptiveBlue, Hakia, Talis, TrueKnowledge, TripIt (an interesting application of t... Read More
On Read/WriteWeb there was a post about 10 semantic Apps to watch. It seems that the terminology of semantics is used for all sort of kinds. Our understanding is that semantic applications should use some form of ontology in order to describe the meani... Read More
Comments
Subscribe to comments for this post OR Subscribe to comments for all Read/WriteWeb posts
Would you call Collexis a semantic app? And Irion? And how about Knowledge Concepts? I would say they fit in your definition, although they are pretty top-down (no social stuff here).
Posted by: Samuel Driessen | November 29, 2007 12:45 AMQuintura utilizes semantic neural networking technology to help users be in control of their web searching and find what they look for much easier.
Posted by: Yakov | November 29, 2007 1:01 AMHi Richard, Check out Imindi.
http://www.imindi.com/help/04What.htm
http://www.imindi.com/journeys/382-semantic-web/maps/3195155-semantic-web
Posted by: Adam Lindemann | November 29, 2007 1:38 AMThere is also an interesting company in the UK called 'Garlik' which does online identity protection with semantic web technology. They even have Tim Berners-Lee in their team.
Yesterday, they launched a new service called QDOS that will give you a web popularity score of sorts. It's actually quite cool. It's hard to describe, but here it is in their words:
"Our digital presence also increasingly opens up new opportunities and influences real world decisions made about us. We now have a means of measuring and therefore managing the way we look online, we call it digital status. QDOS is how we measure it. QDOS is a mirror that reflects your presence in the digital world."
Posted by: Carlos | November 29, 2007 2:31 AMHi Richard,
Nice post and a great round-up. I thought it might be useful to mention that as well as providing a platform for others to build applications, here at Talis we are developing our own applications against our platform.
The first example of this would be Talis Engage, which uses a combination of SKOS, FOAF and SOIC to create a community CMS. There's some more information at http://blogs.talis.com/engage.
Engage was developed using the same APIs available to any Talis Platform developer - so is a good example of what can be achieved.
We've got lots more under development, watch this space!
Chris
Posted by: Chris Clarke | November 29, 2007 2:35 AMHi Richard,
Our approach at Imindi (An application built by a team of Phd computer and neuro scientists) is to take a "mind" approach to helping people to construct their own "mind maps" of connected thoughts and information on any subject. These Mind Maps are little semantic webs that work "As we think" and the "Thought Engine" (Semantic Graph) at the core of Imindi enable Like Minded people (Social Graph) to connect and combine the Thoughts, Information and even Create Knowledge. At the core, all the semantic linkages from every body`s public mind maps collape on themselves to form essentially one global mind map. The closest vision to what we have built was not that of the Semantic Web, but that of the Memex by Vanevver Bush - but as we build a layer of thought and meaning over the information on the web...I would suggest that we are also very much a Semantic App worth watching.
http://www.imindi.com/journeys/382-semantic-web/maps/3195155-semantic-web
Posted by: Adam Lindemann | November 29, 2007 3:59 AM10 Semantic Apps to Watch : i never tried out any of these..
Posted by: TechDune | November 29, 2007 5:32 AMVery interesting list. I have to admit, I hadn't heard of most of these apps. I will be checking them out.
Posted by: Matthew Griffin | November 29, 2007 6:38 AMI've heard of a few them, particularly the search engines, though I'm not so sure if this type of technology would be suited for my own personal searches.
Posted by: Corvida | November 29, 2007 8:20 AMZotero uses both the bottom up and top down approaches to extract semantic information from a very wide range of sites.
Posted by: Dan Cohen | November 29, 2007 9:27 AMNice list. I've been covering various semantic approaches to search from a U.S. Intelligence Community viewpoint. I've added your list to my site at http://www.intelfusion.net.
Posted by: Jeff Carr | November 29, 2007 9:47 AMOntology, OWL & RDfa experiments at http://ontologyonline.org
Posted by: onto | November 29, 2007 9:56 AMHi Richard
Here's one you missed for sure.
The fully Sem Web compliant software, Celtx (www.celtx.com), is a media application that employs a RDF data model. It is used by over 100,000 media creators - so despite the claims of some, it is, as far as we know, one of, if not the only, Semantic Web application in actual, real word use.
We are also building a product called Tadpole. It is a SDK for the rapid deployment of Semantic Web complaint software.
Thanks
Posted by: Mark Kennedy | November 29, 2007 10:04 AMTrue Knowledge and their demo is really interesting. They are a:
Posted by: geekr | November 29, 2007 10:08 AM* Question-Answering Site
* Enhanced Search Engine
* "Wikipedia for Facts"
* Universal Database
* Platform for Building Knowledge Services
Wow: the "semantic" space is dragging hard. Looks like a collection of features (people search and itinerary parsers) and buzz (how long have we heard "natural language search is coming?!") and nebulous blue-sky concepts (freebase??). Not much in the way of actual paradigm shifting real world products here.
After the successes of web2.0, everyone is eager to find the 3.0 and be the next Facebook; IMHO the semantic web has gotten saddled with the 3.0 mantle about 10 years too early.
Some will argue that my perspective lacks vision, but I prefer to think that I'm being pragmatic.
Posted by: rod / techwatching.com | November 29, 2007 10:14 AMwow, those are really the top apps to get installed on my pc. Thanks for the top 10. I liked specially the (HAKIA)
Eliena ANdrews
Posted by: Eliena andrews | November 29, 2007 10:15 AMhttp://healthtreatments.blogspot.com
Does this count as real-world Semantic Web app functionality; i.e. integrating it into architectural software?:
http://www.tenlinks.com/news/PR/graphisoft/112807_tno.htm
(disclosure: I represent Graphisoft)
Posted by: Steve Hodgdon | November 29, 2007 10:48 AMI've been using tripit and loving it. Very easy to figure out and use and keeps everything in one place for me. I do quite a lot of traveling and it has saved me having to compile this information into a word doc.
Great list! Thanks!
Posted by: Chelle | November 29, 2007 10:52 AMHi Richard, great review and excellent round-up! Thanks for including AdaptiveBlue in the list.
I shared some quick thoughts on our blog about the post, AdaptiveBlue, and the exciting future for the web.
It's exciting to see that so many readers haven't heard about these companies - from an innovation perspective all of these firms are providing exciting innovations to digest.
And from a commercialization perspective it'll be interesting to watch all of these players attempt to deliver end-user utility from the innovations. There's definitely a lot of interesting potential.
Posted by: Fraser | November 29, 2007 10:54 AMI've heard a lot of hype about this "semantic web" stuff. Honestly it sounds like "data mining" with a new name. Instead of hyping products that are a long time coming (like natural language search), why doesn't someone work on something that CAN be done NOW, like rearranging search engine results based on which ones the most users clicked? This would be relatively simple to do, and wouldn't be much more vulnerable to abuse than search engines already are, especially if the people administrating the search engine are going to take the time to tinker with their results every so often - I'm looking at you, Google.
Posted by: John Q. Public | November 29, 2007 11:10 AMI work for a company called Reveltyix and we have built and deployed a semantic wiki at www.knoodl.com It has been up and running for a little over a year. It is designed to be an ideal place for communities of interest to collaborate over ontologies because, as anyone that knows semantics should know, any good ontology must be the product of collaboration. We are looking to make significant upgrades in the near future, including adding the ability to translate text documents into OWL ontologies and switching to a much more powerful triple store. The site is free and requires no downloads so feel free to check it out. If you have any questions about the site, you can contact me at mikelangjr@revelytix.com
Posted by: Michael Lang | November 29, 2007 11:15 AMRichard,
Another very exciting and popular Semantic App is ZoomInfo.com . ZoomInfo has close to 40 million people profiles on its site, the majority are generated through Natural languages Processing. Essentially they read webpages and than compose resume style profiles for people. It is a very interesting site and has quite a large user base in the b2b world.
Your on there with quite a few profiles. You can sign up and consolidate yourself. They even have a picture of you!
Posted by: Rob | November 29, 2007 11:21 AMHi Richard,
Very nice roundup. It's great to see all the innovation happening in this area.
Many of the applications you mentioned focus heavily on the connection aspect -- automatically drawing inferences for users for specific applications. We've been working on a different aspect of the semantic web with our data browser, Strata:
http://www.kirix.com
We hope to make it easy for people use the information they encounter in an ad-hoc way, whether the data resides on the web, on their corporate servers, or in local files. Our approach has been assign tabular structure to data such as HTML tables, CSV, RSS, and ATOM, without inferring additional meaning, so users are free to make connections between different data sets as they see fit.
Not a "classic" semantic app (if that is possible at this point :) ), but it does tend to play in the same world by approaching semantic data from a usability angle.
Posted by: Ken Kaczmarek | November 29, 2007 12:11 PMHi Richard --
Posted by: Marian | November 29, 2007 7:23 PMAnother company that should be on your radar in terms of the power of the Semantic Web is Rearden Commerce. The Rearden Personal Assistant is a great example of a Semantic app because it intelligently searches more than 137,000 merchants in its network and finds the most relevant result based on each user’s personal preferences, previous searches, current location and spending policies. There are more than 1,000 businesses, from Fortune 500 to small and medium sized businesses, using the Rearden Personal Assistant to date. You can check out Rearden Commerce at www.reardencommerce.com.
Hmm..but aren't all these "spiffy" apps making our web-life far more bandwidth dependant? The need for bandwidth is ever increasing in times like these, and the supply is nearly not enough. With the innovation and subsequent application of byte-intensive semantics like these, we might just cause the poor (in terms of alloted bandwidth btw) to become poorer and the poppers never catching up with their symmetric twins :(
Unless we look into miniaturization and conservation of the web structure instead of blatantly expanding its ritz, we are up for one stormy night (or century to be precise)
Posted by: Naser | November 30, 2007 7:05 AMIt's still very young (development is really just starting to crank up) but OpenQabal is using Semantic technologies to unite the domains of collaboration, communication, social networking and knowledge management. When it's ready, OQ will be a federated, decentralized social-networking and collaboration platform that enables semantic apps - and integrates with existing knowledge sources and communications facilities - both within an organization as well as across organizations.
Or if you want to call it a Social Operating System, go ahead, I won't try to stop you. ;-)
Posted by: Phillip Rhodes | November 30, 2007 7:06 AMHi Richard,
Our company's efforts should be added to you list of semantics applications.
ThoughtExpress that leverages Orfeo technology uses deep semantic expression to deliver its value. In the systems you mentioned only simple semantics are expressed. For example can you run insurance company only on RDF OWL etc? Of course - no. They have insufficient expressive power to do so. We have found entirely new form of expression for semantics where this is possible as well as doing what RDF etc do. We have found mathematical formulations for knowledge/semantic expressions as well as found difference definitions of next level of things being wisdom. We are working on some very strange but interesting operators in wisdom space. The language follows natural human thought patterns. Anyway, comming back to practical applications of all this, we have entire enterprise management software that uses this semantic language. It is so simple to use that domain specific experts (such as actuaries) can express themselves about their enterprises directly themselves and the whole enterprise - no programming is involved. So far we have developed a network of 12000 concepts - some generic, some country specific, some enterprise specific etc. We are running 4 insurance companies using this technology in on site model. ThoughtExpress will allow all this as a service on the net. To enable this it will come with its own dedicated computing cloud and global process exchange. After seeing the development of facebook, we have also decided that the highest level desk top is not at all what Microsoft has but rather something like FaceBook so we are busy incorporating this kind of desk top. Since, the users will be able to express their enterprises/processes or personal knowledge and processes in the system we are thinking of this as semantic/process WebOs.
Pawel Lubczonok
Posted by: Pawel Lubczonok | November 30, 2007 7:57 AMYet another semantic application to watch may be my Artificial Memory, an semantic (personal) knowledge management system.
Posted by: Lars Ludwig | November 30, 2007 7:59 AMGreat to see these innovative apps here in one place. Much is to learn from this.
What I miss is the role that known technologies play in this area. I'm specifically talking about companies like Autonomy.
They are really into meaning based computing.
Why are the companies you mentioned in the top apps for semantic web and Autonomy isn't?
In my opinion (and part of yours) semantic web is about understanding the meaning of information.
Posted by: Edwin Stauthamer | November 30, 2007 3:08 PMThat is exactly what Autonomy is all about.
Not to mean but:
"Danny Hillis of Freebase (who is as much a tech legend as Berners-Lee)"
If that were true you wouldn't have to tell us!
Posted by: Clyde Smith | December 1, 2007 5:51 AMLifestrea.ms (in private beta now) also claims to be a semantic webapp of importance, as they state on their homepage: "lifestrea.ms stores information about what you read, what you post and what you communicate, we are able to enhance this data semantically and generate a stream of attention data for each of your profiles from this. That way you have a continuously updated APML (Attention Profile Markup Language) file on each of your profile pages."
Posted by: Peter Bekel | December 2, 2007 6:07 AMOrchestr8 AlchemyPoint is another semantic web application worth checking out.
AlchemyPoint is a semantic web / mashup platform that can automatically extract 'context' from the websites you visit, overlaying semantically-relevant information and web links. For instance: if you're looking at a book online, AlchemyPoint detects this and offers all sorts of suggestions (other books by this author, this publisher, prices for this book elsewhere, etc etc etc.)
This system is unique in that it provides users with the ability to create "semantic web mashups" that integrate content from multiple sources, leveraging content like microformats, RDF, RSS/ATOM, etc.
AlchemyPoint also provides a comprehensive web content scraping facility that enables the conversion of unstructured HTML content into a variety of other forms, including XML and even RDF triples!
Posted by: Elliot | December 2, 2007 6:57 AMI'm rather amused by Mark's comment about Celtx : "despite the claims of some, it is, as far as we know, one of, if not the only, Semantic Web application in actual, real word use."
Apparently it's a media application, er, not unlike Microsoft Interactive Media Manager - which is built on RDF & OWL.
Posted by: Danny | December 2, 2007 11:53 AMHi,
Posted by: Nicolas Steegmann | December 3, 2007 6:56 AMCheck Exalead Wikipedia Search which extracts locations, famous people, organizations, statistically related terms and Wikipedia categories out of every Wikipedia article.
Hi,
Posted by: Daniel | December 3, 2007 10:03 AMAnother interesting semantic application has been build in Switzerland. The available solution for news aggregation and navigation is running under http://news.ontos.com. The solution is using NLP technology to automatically annotate named entities and also the relations between the named entities. Compared to Gnosis it has also the relations which gives an extended view and allows better classification and filtering. In the back the systems is based on standards like RDF and OWL.
Hi all,
Posted by: Roberto B. | December 7, 2007 12:43 AManother news aggregation service available at Libero Portal (Wind Telecomunicazioni S.p.A) uses text mining techniques to analyze the content of the Italian news articles processed in Libero WebNews (from >1000 Italian newspapers), with the goal of highlighting the main subjects, i.e. the people, the products, the localities, the societies, the institutions, etc., and extracts the relations that elapse between them. The final result of this elaboration, GraphNews,
is a graph visualization of these subjects and their relations.
The graph is browsable through a click on a node (ellipse box) where a new graph with subject and relations tied to that specific node is showed, or a click on an arc (square box) where are returned all the news concerning about the two subjects joined from the arc. GraphNews also allows the change of detail level and the time period (day, week, month) of the graph.
Bighow.com connects Community Publishing with Semantic-like categorizations.
Posted by: Pramit | December 8, 2007 12:47 AMCheck it out and you will understand what I mean.
Localina is a location recommendation system. You rate the locations you already know and you will get new recommendations.
For giving recommendations, Localina needs your ratings! As soon as Localina is 100% loaded, recommendations are possible.
The recommendation system will be based on your profile, your ratings and your contacts. Currently they are in private beta. Did not take long time after registration until an invitation code was sent. They will add events soon... www.localina.com
Posted by: Marco | December 11, 2007 4:50 AMExciting to see all these new startups coming up! Interesting to note however, that are quite some established players particularly on the European market, like moresophy, that offer semantic platforms. These are are the base for more and more REAL enterprise applications implemented by business partners.
The platform is - IMHO - the only one available on the market that is able to flexibly handle context in the same flexible manner that humans are doing when they are reflecting issues from different perspectives.
Posted by: Heiko Beier | December 19, 2007 10:52 AM