Increasing Relevancy and Trust on the Web: Powder Adds Steps in the Right Direction

Constant progress is being made to disambiguate the information mine on the Web by using semantic web metadata to mark up web content, using RDF, OWL, and HTTP, but with additional features which add more descriptive metadata to resources. The POWDER (Protocol for Web Description Resources) W3C Working Group is behind the initiative, and major Web players are starting to notice. A number of developments were published on the 15th of August 2008, including Formal Semantics, Grouping of Resources, Description Resources, Primer, and Test Suite.  The work of POWDER contributes to one overarching objective: improving both the developer’s and the user’s experience of the Web, from a number of perspectives. These can be stated in terms of three related concepts: making the Web more personal, relevant and trustworthy. Given the fundamental role of these characteristics in search (engines), this should be precisely the reason why global search companies are displaying interest. For example, Powder’s most recent Outreach day, on the 22 September, was sponsored by Yahoo! 

Powder enhances Web content by increasing its relevance, as each piece of content is marked up, specifying what it is in terms of RDF. So, this is already known by most Semantic Web developers. What sets POWDER apart is the inclusion of a number of other features, such as classifying single items within content groups, Data Retrieval Efficiency, allowing the user to decide whether a resource (such as a PDF) is relevant for their requirements before they download it, based on the description, profile matching, trustmarks, and semantic annotation. Profile matching enables the user to judge whether the resource which will be retrieved is actually what they are looking for, in terms of their preferences, for example, is it mobile compatible?, does it have child protection? Trustmarks allow partners in the web to verify the claims made by the descriptions and act as certification authorities, and the final feature is easy semantic annotation, effectively, putting the pow(d)er in the user’s hands.

 http://www.w3.org/2007/powder/

Tags: , , , , , , , ,

Related posts

The Semantic Web: Its Status in the IT Industry

David Provost, a business development professional and a strategist, published a report on the 30th of September which outlines a global review of the current position of the Semantic Web in the Information Technology industry, providing a commercial slant on the advancement of the Semantic Web in IT companies. Initially, some key concepts of the Semantic Web are summarized, providing both less-technical entrepreneurs and those on the development side with a general overview of the Semantic Web. The executive summary of the article mentions some key emerging trends: linked data, Social Network Analysis and the Semantic Web’s role in this scenario, and the role played by Natural Language Processing (NLP), Semantic search and the traditional publishing industry. However, most significantly, this summary captures an evident trend which has recent become more and more notable in the industry: the transition from technical descriptions of technology, to an emphasis on problems, solutions and applications.

Given that new companies which rely on Semantic Web technology are continuously emerging, it appears that the moment has arrived for these IT businesses to differentiate themselves in the market. This report provides a step in this direction, by providing a summary of what these leading companies, or vendors, are comprised of and what they offer. The information is based on phone or face-to-face personal interviews, with representatives of 17 companies, analysed for credibility with regard to being a “going concern”, in other words, an active companies with substantial cash flow (according to the author; those companies who have used spare cash to sponsor a conference, however, this is not a definitive indicator of the financial status of an organisation). Ten companies were based in North America, six in Europe, and one in South Korea.

These organizations address diverse topics within the Semantic Web, such as: databases for storing semantic data, ontology engineering and management tools, application development platforms, and NLP. One important transformation is that previously, companies just sold software. Now, the transition to “do” solutions rather than “make” solutions is evident, with vendors selling ready-made solutions with less emphasis on the development side. For example, solutions in areas such as: knowledge management, risk management, and content management, among others. The author redefines this trend as a change in market strategy: companies have widened their market, targeting companies who previously could not afford to employ complex development skills to adopt a solution to their needs. Now these companies are being offered solutions which already suit their internal requirements, without having to make substantial modifications.

A diagram is provided, which lists the following 17 companies as having taken part in the review: Aduna, The Calais Initiative, Cambridge Semantics, Dow Jones Client Solutions, Expert System, Franz, Mondeca, Ontoprise, Ontos, OpenLink Software, Primal Fusion, Saltlux, Sindice, Thetus, TopQuadrant, Twine/Radar Networks, and Yahoo!/Searchmonkey. The diagram marks each company schematically according to the functions they possess, defined as the following: Solution, Middleware, NLP, Database, Platform, Ontology, Search, Consumer Web Service, and Developer Web Service.  Solution and Platform exhibit the highest frequency; Developer Web Service and Consumer Web Service the least.    Subsequently, the findings of the report are published, in the form of four key trends: Semantic Web technology has emerged to be fierce competition for companies providing solutions using traditional technology, NLP has transpired to be a fundamental tool for mining content for the Web, the value and use of Linked Data is slowly gaining recognition, and the role of marketing, technical and solution partners is becoming essential in selling Semantic Web based solutions.

The principal findings are followed by a profile of each of the organizations interviewed, defined in terms of their products, employees, revenue, Installed base, primary offering, key differentiators, six/twelve month plan, and a final analysis. Given that the article has been written from a strategist’s perspective, it should be viewed as such, as it focuses on the aspects for the marketing of companies, rather than any shortcomings. It is not an entirely objective review, and some companies, such as Talis, are missing, but it is nevertheless an interesting read and a good overview.

http://pdfmenot.com/view/http://www.davidprovost.com/Resources/Semantic%20Web%20Industry%20Revie.pdf

Tags: , , , , , , , , , , , , , , , , ,

Related posts

Band Metrics: Semantic Web Technology for Gathering and Analysing Opinion Trends of Musicians and Bands

Yesterday a new take on the use of Semantic Web Technology in the music industry emerged: Band Metrics announced the launch of their private beta on their blog, which helps musicians and bands manage their digital identity, as well as monitoring bands’ popularity and analysing trends. Currently, more details regarding the technology are not available, as it is patent pending. Band Metrics is a TechCrunch50 semi-finalist and will participate in its DemoPit, where it will present the demo tomorrow.

 http://www.techcrunch50.com/2008/conference/demopit.php

http://www.bandmetrics.com/beta/#

Tags: , , , , ,

Related posts

Mozilla Labs Release Ubiquity: A Firefox Plugin Which Responds to Real Language

Last week witnessed an important release by Mozilla Labs: a prototype of a Firefox extension which actually responds to a users request rather than a search query, using a command line. It is one of the first effective examples of a plugin which transcends the boundaries of traditional search: for example, if a user wishes to send an email to a friend arranging to meet in a restaurant, he no longer has to search separately for links to a map of the location, restaurant reviews, and open his email client: Ubiquity performs this process in a single step. Tutorials and examples of Ubiquity can be found on http://labs.mozilla.com/2008/08/introducing-ubiquity/

Tags: , ,

Related posts

Details of a New Semantic Search Engine and Health Knowledge Base Released by WebLib


This morning the development of a new semantic search engine and health knowledge base was announced by the company WebLib, experts in search solutions based on Natural Language Processing and Semantic Web Techniques. The solution claims to provide search features currently lacking in engines which retrieve health-related information: relevant, current, and actionable. The search will be powered by the Healthmash Health Knowledge Base, and will combine Web 2.0 technologies with Semantic Web methods to retrieve user-relevant information. However, a beta version is not yet available, it will become available at the end of 2008.

http://www.prweb.com/releases/health_knowledge_base/200808/prweb1226424.htm
http://www.weblib.com/

 

Tags: , , ,

Related posts

Twine Creator Nova Spivack Outlines his Vision of the Future Desktop

Yesterday the founder of the leading semantic collaboration service Twine published his description of the evolution of the desktop beyond traditional boundaries. This post discusses some of the principal features of what some have coined the “Webtop”. The fundamental question is that as many applications are emerging for Web-hosted storage, both for individuals and enterprises, what will the desktop of the future look like, and which features will ensure its existence?

It is exactly due to this question that Spivack commented that this desktop has “yet to be invented”. One mayor feature which will underlie this desktop, whatever its form, is that the user will still have a personal workspace, but it will not live statically on one device. It will be available on any device the user connects to, being part of a mobile cloud. Alongside this characteristic, it is a natural evolution that the browser, rather than the traditional desktop, will become the central feature. If users ubiquitously log on to a number of devices during the day, the desktop will be literally launched from the browser, instead of the other way round.

There will be a focus on updating information instead of organizing it. Nowadays, the key to keeping informed is having timely access to information, rather than archiving it in directories. This trend is reflected in social news sites and RSS feeds. This will be accompanied by users being in control of which information they absorb, as this capacity is finite. The user must decide what to pay attention to and what trends to follow; effectively, the mindset of a day trader.

Another pattern which has previously been pointed out by Tim Berners-Lee is that of “collective intelligence”, the future desktop will be characterized by a social aspect, being part of a social community which aids information management. Semantic, structured search being provided by the new desktop will also help the user in managing information. The social dimension of the desktop introduces many other useful features, such as social search, for example, according to document rankings assigned by members in a user’s network. Interactive shared spaces will replace folders, where many contributors can work collectively on one problem. The future desktop will be portable, being characterized by open-standard data formats. This will need to be accompanied by a model for policies and permissions regarding data, in order to avoid privacy concerns and stimulate mass adoption.  

Not only will the future desktop be online, attention-focused, social, portable, and have search capabilities, it will also be smart. The presence of the aforementioned features creates a desktop which is an intelligent system, relieving the user of information handling tasks. That means, adopting to the user’s personal preferences; interests, activities, work requirements, in a timely manner. It will be a personal cloud supported by distributed services, viewed as if they were all in the one space. That leaves only one question: who will be the first to propose a WebOS?

http://www.twine.com/

http://www.readwriteweb.com/archives/future_of_the_desktop.php

Tags: , , , ,

Related posts

Sir Berners-Lee’s Insights on the Future of the Semantic Web

An important public symposium in the Semantic Web field was held last week in Rensselaer Polytechnic Institute (NY, US), as part of the launch of a new research institute for Web Science, entitled Tetherless World Research Constellation. A number of leading figures such as Nova Spivack and Nigel Shadbolt were present, among those Tim Berners-Lee. The principal focus of Berners-Lee’s keynote speech was his future vision of the Web, for which he discussed a number of distinct aspects. Berners-Lee has never faltered to advocate his future Web vision as a semantic and social one. The formal scientific analysis of such outlooks has emerged as the field of Web Science, to which the new research centre is dedicated. The characteristics of the evolving Web which Berners-Lee outlined could be divided into various areas. One of the topics, which was also one of the general themes of the day, was the construction of more intelligent data, as opposed to smarter software. This is one of the objectives of the Semantic Web, which was introduced by Berners-Lee, however, adding that the conceptual links between data have the power to be used in unpredictable and novel ways. It has not yet emerged what the outcome of such opportunities will be, but the evolution of the Web is likely to have an impact on the traditional methods of constructing social systems. This was another one of the topics elaborated by Berners-Lee and other attendees.

The Web has spawned new social systems which have initiated new possibilities for viewing science and other political systems such as democracy. It is evident that with the advent of the Social Web, that networks can influence collective thinking, ideas or movements, which may be constructive but also destructive. The effect of such systems will be witnessed in the future.

 A further topic discussed was the requirement for existing technology to be able to cope with and complement the future form of the Web. If data and the concepts contained in data are interlinked, there arises a requirement for technologies to adapt to this data through more pages, higher bandwidth, and mobile devices. The filtration of the Web into daily life is an ongoing theme, which has previously been discussed by many of the leading researchers in the Semantic Web domain, which advocate and emphasize the role of the mobile Web. For example, the chairman of the Mozilla Foundation, Mitchell Baker. 

 http://www.forbes.com/feeds/ap/2008/06/11/ap5106902.html

http://www.pcmag.com/article2/0,2817,2319807,00.asp

Tags: , , , , , , ,

Related posts

A Medley of Semantically Related News Snippets


Today witnessed a number of thought-provoking news publications from various sources, from a newspaper in Madrid to ReadWriteWeb. My first observation this morning, in a newspaper circulated in Madrid’s metro, is worth mentioning, as it demonstrates the triumphant transition made by Web 2.0 into the general public’s eye, as well as recognising the need for technologies to aid users in activities such as managing their social profiles and their blog profiles. The daily cartoon sketch shows two female students seated in a university computer lab. One asks the other: “How’s it going?”. Her friend answers: “Ugh……..I’m updating my social profile in 18 social networks, changing my website’s photos, and renewing the contents of my blog”. The response is: “And on top of all that, they expect us to study”. Although the underlying issues provoked in this cartoon have been partly resolved by initiatives such as Google Open Social, there is still significant room for improvement.

Besides the spine-chilling news for Google that Powerset unleashed its test version of a semantic search engine for Wikipedia for public use on Monday (which I will not go into here, as there are currently countless articles floating around the Web discussing what some have coined the “Google Killer”), another search engine, entitled Uptake, was launched today. Uptake, formerly known as Kango, is a travel search engine which extracts information from more than 1000 travel sites in order to construct a database of over 400,000 US hotels and activities. Uptake has built its database from consumer reviews, opinions and descriptions on these sites, and has constructed an ontology from metadata applied to the content of these sources. One of the more recent Natural Language Processing Techniques Uptake applies is Sentiment Analysis, also referred to as Opinion Mining, which uses syntactic parsing to extract words to indicate, for example, favourable sentiment towards a hotel, such as “good time”, “fantastic view” or “relaxed atmosphere”, and distinguishes positive sentiment from negative sentiment.

Today also saw the move of Jeremy Carroll, lead architect on the Open Source Jena Toolkit at HP, to TopQuadrant, a leading Semantic Web company, as Chief Product Architect.

Tags: , , , , , , ,

Related posts

Semantic Technologies: Evidence in Business

The success and resulting growth of semantic technologies in corporate scenarios for a multitude of business planning and development tasks is becoming increasingly evident. Two important upcoming conferences on Semantic Web technologies will unveil and discuss the most recent technologies among such achievements. The 2008 Semantic Technology conference will be held from May 18-22 in San Jose, California, whose principal objectives include demonstrating Semantic Web technologies in a number of domains: government, healthcare, finance, semantic SOA, application development, and the TopBraid suite. Today, a press release revealed that 16 significant new semantic innovations would be exhibited at SemTech, including novel applications in Natural Language Processing and Text Analysis, Enterprise-wide Application Deployment and Integration; Data Searching, Sharing, and Tracking; Knowledge Modeling Solutions; Information Publishing and Data Provisioning; Open Source Community Initiatives; and Artificial Intelligence Functionality, in particular, a computer chip which claims to behave like a human brain by performing the behavior of a human brain: the capability to reason, think and analyze. Another important conference follows the SemTech conference – the LinkedData Planet Conference and Expo, which will be held in New York City on June 17-18, whose keynote speakers include Kingsley Idehen, President and CEO, OpenLink Software Inc., Ian Davis, Chief Technology Officer and Director, Talis Group, Sir Tim Berners-Lee (no introduction needed…), Atanas Kiryakov, Head, Ontotext Lab (Sirma Group) , and Dr. Anant Jhingran, VP and CTO, Information Management Division, IBM. The aim of the conference is similar to that of SemTech, to provide industry professionals and technologists with new views and applications for organizational data processing and business problems, such as trend analysis, decision support systems, social networking, and supply chain relationship management.  http://www.semantic-conference.com http://www.linkeddataplanet.com 

Tags: , , , , , , , , , ,

Related posts

Semantic Labeling According to Conceptual Connections

The startup company InfoGenome adds some novel “out-of-the-box” insights to the process of adding metadata to the overwhelming volumes of information a user is faced with every day. On May 5, ReadWriteWeb published a description of InfoGenome’s i360 tool, which, in striking contrast to other semantic applications, adds human knowledge to data rather than extractable RDF. The only input i360 requires from the user is to “drag and drop” relevant information, such as a link, into the application and add some text as its description. This step is referred to by the company as “fire and move on” Other connections are made organically and rapidly by the user, by dragging and dropping any further data relevant to the link into the application, as it is encountered over time. Any entities such as people or places associated with the information already entered can be added by the user. This concept is contrary to the long-standing idea of the application of algorithms for organization of data.
The conceptual idea captured by the method is that instead of humans having to remember associations between distinct data in different formats, the software remembers the user’s mental connections between data, for humans. So, this company has discovered a new marketing idea – focusing on the power of the human brain, by organizing info according to how an individual user really constructs it conceptually. Which is not machine-readable.
So, if it is not machine-readable, why is it useful? On the i360 website, the utility of the application is presented in the context of a corporate scenario, claiming to generate significant productivity payoffs. For example, Information Workers in the enterprise are Team Members which congregate significant “Information Insights” in their human memory. These insights can be saved by i360, and can thus be retained in the enterprise when a Team Member leaves, such as free connections between any elements encountered during a company mission. For example, associating emails from various colleagues to different companies, such as Microsoft or Google. This organizes the information as the user requires it to complete his mission.
http://www.infogenome.net

Tags: , , ,

Related posts