Artificial Intelligence Lab, Department of MIS, University of Arizona / Digital libraries of geo-spatial multimedia content are currently deficient in providing fuzzy, concept-based retrieval mechanisms to users. The main challenge is that indexing and thesaurus creation are extremely laborintensive processes for text documents and especially for images. Recently, 800,000 declassified satellite photographs were made available by the United States Geological Survey. Additionally, millions of satellite and aerial photographs are archived in national and local map libraries. Such enormous collections make human indexing and thesaurus generation methods impossible to utilize. In this article we propose a scalable method to automatically generate visual thesauri of large collections of geo-spatial media using fuzzy, unsupervised machine-learning techniques.

Digital Library for Earth Science Education, DLESE / The community digital library has emerged as a recent evolution in the development of digital libraries. A community digital library is distinct through having a community of potential users define and guide the development of the library. In this article, we present how the geoscience community has worked to develop The Digital Library for Earth System Education (DLESE) in the light of recent work in the area of meta-design, the design of the design process. The development of DLESE is described utilizing a conceptual framework developed from the analysis of a variety of open source projects.

The Belgians Otlet en La Fontaine created the Universal Decimal Classification in order to collect and organize the world's knowledge. This happened in an age when information was almost exclusively made available by libraries. Since the internet, the quantity of information outside libraries is enormous and keeps growing every day. The internet is accessible to anybody, it is fundamentally unorganized and its content changes constantly. Collecting and organizing the world's knowledge seem to have become an impossible ambition. Perhaps it is even unnecessary, since search engines make information retrievable now. And why would we organize information if we can find it? So what will be the role of UDC and libraries in this internet environment? Libraries can still play a role as a major information provider, if they adapt fully to the expectations of a modern end user. The design and the functionalities of online catalogues should allow maximal accessibility, usability and active participation of the end user in the internet environment. Metadata, like UDC, should maximize the visibility of information, enrich it and invite the end user to assign metadata himself.

Digital Libraries and Digital Repositories are data-intensive with large numbers of fulltext resources accessible online. Activities in the area of Semantic Web development recognize the significant part played by metadata and knowledge organization systems such as classification systems and thesauri in capturing and communicating ‘meaning’. We now have Web ontology standards, such as Simple Knowledge Organization Systems (SKOS), a common data model for sharing and linking knowledge organization systems via the Semantic Web. Standards such as SKOS are also meant to be used as a vehicle for deployment of knowledge organization systems that were not born digital (or XML/RDF) such as thesauri and bibliographic classifications. This paper attempts to present an application of the faceted classification scheme as enunciated by Ranganathan in developing ontologies. It further explores the issues in modelling the faceted scheme of Ranganathan using SKOS.

Digital heritage systems usually handle a rich and varied mix of digital objects, accompanied by complex and intersecting workflows and processes. However, they usually lack effective workflow management within their components as evident in the lack of integrated solutions that include workflow components. There are a number of reasons for this limitation in workflow management utilization including some technical challenges, the unique nature of each digital resource and the challenges imposed by the environments and infrastructure in which such systems operate. This thesis investigates the concept of utilizing Workflow Management Systems (WfMS) within Digital Library Systems, and more specifically in online Digital Heritage Resources. The research work conducted involved the design and development of a novel experimental WfMS to test the viability of effective workflow management on the complex processes that exist in digital library and heritage resources. This rarely studied area of interest is covered by analyzing evolving workflow management technologies and paradigms. The different operational and technological aspects of these systems are evaluated while focusing on the areas that traditional systems often fail to address. A digital heritage resource was created to test a novel concept called DISPLAYS (Digital Library Services for Playing with Antiquity and Shared Heritage), which provides digital heritage content: creation, archival, exposition, presentation and interaction services for digital heritage collections. Based on DISPLAYS, a specific digital heritage resource was created to validate its concept and, more importantly, to act as a test bed to validate workflow management for digital heritage resources. This DISPLAYS type system implementation was called the Reanimating Cultural Heritage resource, for which three core components are the archival, retrieval and presentation components. To validate workflow management and its concepts, another limited version of these reanimating cultural heritage components was implemented within a workflow management host to test if the workflow technology is a viable choice for managing control and dataflow within a digital heritage system: this was successfully proved.

Thesis (Ph. D.)--University of Texas at Austin, 2002. / Vita. Includes bibliographical references. Available also from UMI Company.

The article addresses practicing librarians and people involved in the creation of digital libraries and institutional or subject repositories by making use of one or more of the many softwares freely available for the purpose. This is an attempt to draw up a checklist of factors to be considered while planning for these projects. The emphasis here is on non-technical, project related issues. An attempt is made to explicitly state and highlight issues that are often overlooked.

InterLibrary Cooperation is an important activity of libraries. However, the emerging digital library architecture has not explicitly supported this requirement. In this paper, we propose an InterLibrary Cooperation Framework for digital libraries using P2P technology. An application of this framework to Faculty Publication Sharing System is presented. Besides, a reputation model based on data mining is utilized to provide libraries with incentives to join this framework.

Artificial Intelligence Lab, Department of MIS, University of Arizona / In this paper we describe MedTextus, an online medical search portal with dynamic search and browse tools. To search for information, MedTextus lets users request synonyms and related terms specifically tailored to their query. A mapping algorithm dynamically builds the query context based on the UMLS ontology and then selects thesaurus terms that fit this context. Users can add these terms to their query and meta-search five medical databases. To facilitate browsing, the search results can be reviewed as a list of documents per database, as a set of folders into which all the documents are automatically categorized based on their content, and as a map that is built on the fly. We designed a user study to compare these dynamic support tools with the static query support of NLM Gateway and report on initial results for the search task. The users used NLM Gateway more effectively, but used MedTextus more efficiently and preferred its query formation tools.

Artificial Intelligence Lab, Department of MIS, University of Arizona / Research shows that recommendations comprise a valuable service for users of a digital library [11]. While most existing recommender systems rely either on a content-based approach or a collaborative approach to make recommendations, there is potential to improve recommendation quality by using a combination of both approaches (a hybrid approach). In this paper, we report how we tested the idea of using a graph-based recommender system that naturally combines the content-based and collaborative approaches. Due to the similarity between our problem and a concept retrieval task, a Hopfield net algorithm was used to exploit high-degree book-book, useruser and book-user associations. Sample hold-out testing and preliminary subject testing were conducted to evaluate the system, by which it was found that the system gained improvement with respect to both precision and recall by combining content-based and collaborative approaches. However, no significant improvement was observed by exploiting high-degree associations.