Stanford University's Knowledge Systems Laboratory (KSL) is working in partnership with Battelle Memorial Institute and IBM Watson Research Center to develop a suite of technologies for information extraction, knowledge representation & reasoning, and human-information interaction, in unison entitled 'Knowledge Associates for Novel Intelligence' (KANI). We have developed an integrated analytic environment composed of a collection of analyst associates, software components that aid the user at different stages of the information analysis process. An important part of our participatory design process has been to ensure our technologies and designs are tightly integrate with the needs and requirements of our end users, To this end, we perform a sequence of evaluations towards the end of the development process that ensure the technologies are both functional and usable. This paper reports on that process.

T he Semantic Web 1 has failed to produce communities, quantity, or quality. Very little public information is presently available in the Resource Description Format (RDF) or Web Ontology Language (OWL) data formats, and Semantic Web services are few and far between. This leaves the existing Web with very little semantic structure. In the last issue (November/Decem-ber 2005, pp. 86–87), I examined the problems facing the Semantic Web effort. Here, I propose a solution to the problem of introducing widespread semantics to the World Wide Web. My proposal is to do for the Semantic Web what Tim Berners-Lee (www.w3.org/ People/Berners-Lee) did for Project Xanadu, the original hypertext project (www.xanadu.com), and Standardized General Markup Language (www.w3. org/MarkUp/SGML/). When Berners-Lee developed the Web, he took the salient ideas of hypertext and SGML syntax and removed complexities such as backward hyper-links. At the time, many criticized their absence from HTML because, without them, pages can simply vanish and links can break. But the need to control both the linking and linked pages is a burden to authoring, sharing , and copying. Similarly, early forms of HTML paid no regard to SGML document-type definitions (DTDs). Berners-Lee simply ignored these difficult to create and understand declarations of how markup tags are used. Semantic Web proponents should take a lesson from this and more recent Web-based successes. Two recent Web communities are worth examining in this regard. The Flickr photo-sharing site (www.flickr. com) has developed a community of people who " tag " their images with keywords for high-quality image retrieval. The system is almost suicidally simple — there's no notion of synonyms or disambiguation — but simplicity encourages participation. From an academic standpoint, the lack of semantics in tags implies that the system should collapse into incoher-ence with even moderate participation. Yet, it is one of the most successful Web communities today. For a Web community with simple, easy-to-use authoring tools that support synonyms, disambiguation, and categories, we can look to Wikipedia, the Internet encyclopedia (http://en. wikipedia.org). It's based on wiki technology , which lets people write HTML in plaintext, as if writing an email — automatically generating bullet points, for example, by starting a line with an asterisk — and supports a very simple set of semantics. Wikipedia calls synonyms redirect pages, and disam-biguation is explicitly handled via special pages. Wikipedia's major limitation is that it's a " shadow " web. It will always …

A central theme of the Semantic Web is that programs should be able to easily aggregate data from different sources. Unfortunately, even if two sites provide their data using the same data model and vocabulary, subtle differences in their use of terms and in the assumptions they make pose challenges for aggregation. Experiences with the TAP project reveal some of the phenomena that pose obstacles to a simplistic model of aggregation. Similar experiences have been reported by AI projects such as Cyc, which has led to the development and use of various context mechanisms. In this paper we report on some of the problems with aggregating independently published data and propose a context mechanism to handle some of these problems. We briefly survey the context mechanisms developed in AI and contrast them with the requirements of a context mechanism for the Semantic Web. Finally, we present a context mechanism for the Semantic Web that is adequate to handle the aggregation tasks, yet simple from both computational and model theoretic perspectives.

The web lacks support for explaining information provenance. When web applications return answers, many users do not know what information sources were used, when they were updated, how reliable the source was, or what information was looked up versus derived. Support for information provenance is expected to be a harder problem in the Semantic Web where more answers result from some manipulation of information (instead of simple retrieval of information). Manipulation includes, among other things, retrieving, matching, aggregating, filtering, and deriving information possibly from multiple sources. This article defines a broad notion of information provenance called knowledge provenance that includes proof-like information on how a question answering system arrived at its answer(s). The article also describes an approach for a knowledge provenance infrastructure supporting the extraction, maintenance and usage of knowledge provenance related to answers of web applications and services.

The web lacks support for explaining information provenance. When web applications return answers, many users do not know what information sources were used, when they were updated, how reliable the source was, or what information was looked up versus derived. Support for information provenance is expected to be a harder problem in the Semantic Web where more answers result from some manipulation of information (instead of simple retrieval of information). Manipulation includes, among other things, retrieving, matching, aggregating, filtering, and deriving information possibly from multiple sources. This article defines a broad notion of information provenance called knowledge provenance that includes proof-like information on how a question answering system arrived at its answer(s). The article also describes an approach for a knowledge provenance infrastructure supporting the extraction, maintenance and usage of knowledge provenance related to answers of web applications and services.

In this paper, we describe TAP, an experimental system for identifying and researching many different of the different technical issues that lie on the path to achieving the vision of the Semantic Web. In particular, we address the issues of scalable query languages, sharing vocabularies, bootstrap knowledge bases, automated extraction of RDF from text and applications of the Semantic Web.