PDF, 4.7 MB
Zipped PostScript, 16.7 MB
PDF, 4.7 MB
Zipped PostScript, 16.7 MB
HTML
HTML
PDF, 504 KB
Zipped PostScript, 911 KB
In this paper we provide an overview of key changes that happened on the Web in a few recent years. We start by analyzing changes occurring at the level of widely understood Web infrastructure (standards, computing, storage). Then, we focus on machine-oriented and user-centric trends in representation of information (both structured and unstructured). Next, we briefly discuss evolution of types of on-line functionalities and their access modes. Fourth component of the Web that we analyze is related to a few directions in actual usage of Web and its impact on social life. Final part of this paper is devoted to topics that span previous components such as driving forces, business models and privacy.
PDF, 303 KB
Zipped PostScript, 1.0 MB
Weblogs are widely known as a technology that allows publishing textual content in reverse chronological order, often expressing the subjective points of view of a single or multiple weblog authors. The simplicity and autonomy of weblogs is assumed to play a fundamental role in their popularity and their ability to transform implicit knowledge into explicit forms. In recent years, enterprises began to experiment with weblogs to facilitate inter- and intraorganizational knowledge sharing. Although weblogs have been increasingly adopted in a corporate context, sound exploratory and explanatory knowledge and theories about weblog adoption practices in corporoate contexts are missing. A rich toolset of network-analytic techniques exists to analyze the vast amount of electronic traces produced by large weblog networks. However, in small and medium enterprises, electronic traces are sparse due to the lack of a critical amount of weblogs being maintained, and weblog communications are intervowen with offline exchanges. This requires researchers to adopt and develop new analytical techniques and concepts for advancing the state of research on weblogs. Our paper is intended to expand existing research on corporate weblogs by studying weblog adoption practices for knowledge transfer purposes in Small and Medium Enterprises. In this paper, we report selected findings from a case study in which a weblog was used to facilitate knowledge transfer in an SME. The overall contributions of our paper are deep insights into a single case of a weblog adoption in a small and medium enterprise and the formulation of a set of tentative hypothesis.
PDF, 805 KB
Zipped PostScript, 1.3 MB
Semantic-based information retrieval is an area of ongoing work. In this paper we present a solution for giving semantic support to multimedia content information retrieval in an e-Learning environment where very often a large number of multimedia objects and information sources are used in combination. Semantic support is given through intelligent use of Wikipedia in combination with statistical Information Extraction techniques.
PDF, 841 KB
Zipped PostScript, 440 KB
This paper studies unevenness in network properties on the social Semantic Web. First, we propose a two-step methodology for processing and analyzing social network data from the Semantic Web. Using the SPARQL query language, a derived RDF graph can be constructed that is tailored to a specific question. After a brief introduction to the notion of unevenness, this methodology is applied to examine unevenness in network properties of semantic data. Comparing Lorenz curves for different centrality measures, it is shown how examinations of unevenness can provide crucial hints regarding the topology of (social) Semantic Web data.
PDF, 1.0 MB
Zipped PostScript, 11.2 MB
Large portions of the Web are buried behind user-oriented interfaces, which can only be accessed by filling out forms. To make the therein contained information accessible to automatic processing, one of the major hurdles is to navigate to the actual result page. In this paper we present a framework for navigating these so-called Deep Web sites based on the page-keyword-action paradigm: the system fills out forms with provided input parameters and then submits the form. Afterwards it checks if it has already found a result page by looking for pre-specified keyword patterns in the current page. Based on the outcome either further actions to reach a result page are executed or the resulting URL is returned.
PDF, 261 KB
Zipped PostScript, 306 KB
Companies use company-specific terminology that may differ from the terminology used in existing corporate ontologies (e.g. Tove) and therefore need their own ontology. However, the current ontology engineering techniques are time-consuming and there exists a conceptual mismatch among developers and users. In contrast, folksonomies or the flat bottom-up taxonomies constituted by web users' tags are rapidly created. In this paper, (1) we present an approach that cost-efficiently derives a lightweight corporate ontology from a corporate folksonomy, (2) by means of a folksonomy dataset from a European company, we provide preliminary evidence that our suggested approach reflects the company-specific terminology, (3) we detect a number of possible applications for the company when implementing the presented methodology on a corporate folksonomy and (4) as an additional evaluation, we asked the company to briefly evaluate the results and possible applications.
PDF, 592 KB
Zipped PostScript, 429 KB
The deep Web has many challenges to be solved. Among them is schema matching. In this paper, we build a conceptual connection between the schema matching problem SMP and the fuzzy constraint optimization problem FCOP. In particular, we propose the use of the fuzzy constraint optimization problem as a framework to model and formalize the schema matching problem. By formalizing the SMP as a FCOP, we gain many benefits. First, we could express it as a combinatorial optimization problem with a set of soft constraints which are able to cope with uncertainty in schema matching. Second, the actual algorithm solution becomes independent of the concrete graph model, allowing us to change the model without affecting the algorithm by introducing a new level of abstraction. Moreover, we could discover complex matches easily. Finally, we could make a trade-off between schema matching performance aspects.
PDF, 774 KB
Zipped PostScript, 1.2 MB
The increasing need of small knowledge-intensive companies for loosely-coupled collaboration and ad-hoc knowledge sharing has led to a strong requirement for an alternative approach to developing knowledge management systems. This paper proposes a framework for managing organisational knowledge that builds on a socio-technical perspective and considers people as well as technology as two highly interconnected components. We introduce a conceptualised system architecture that merges enterprise social software characteristics from the realm of Enterprise 2.0, and information processing techniques from the domain of Semantic Web technologies. In order to deliver a KM approach that could assist in reducing the socio-technical gap, we suggest deploying such a solution using an integrated sociotechnical implementation methodology.
PDF, 536 KB
Zipped PostScript, 1012 KB
The increase of digital bandwidth and computing power of personal computers as well as the rise of the Web 2.0 came along with a new web programming paradigm: Rich Internet Applications. On the other hand, powerful server-side business rules engines appeared over the last years and let enterprises describe their business policies declaratively as business rules. This paper addresses the problem of how to combine the business rules approach with the new programming paradigm of Rich Internet Applications. We present a novel approach that reuses business rules for deriving declarative presentation and visualization logic. In this paper we introduce a rule-driven architecture capable of executing rules directly on the client by implementing the Rete algorithm. We propose to use declarative rules as platform independent language describing the application and presentation logic. By means of AJAX we exemplarily show how to use client-side executable rules for adapting the user interface of Rich Internet Applications. We call our approach ARRIA: Adaptive Reactive Rich Internet Applications. In order to show the usability of our approach we explain our approach based on an example taken from the financing sector.
PDF, 263 KB
Zipped PostScript, 394 KB
A trust metric is a technique for predicting how much a user of a social network might trust another user. This is especially beneficial in situations where most users are unknown to each other such as online communities. We believe the recent tumultuous evolution of social networking demands for a collective research effort. With this in mind we created Trustlet.org, a platform consisting of a wiki for open research on trust metrics. The goal of Trustlet is to collect and distribute trust network datasets and trust metrics code as Free Software, in order to facilitate the comparison of different trust metrics algorithms and a more coherent progress in this field. At present we made available some social network datasets and code for some trust metrics. In this paper we describe Trustlet and report a first empirical evaluation of different trust metrics on the Advogato social network dataset.
HTML
