Vol. 28 · No. 6

A growing number of users in Web 2.0 based social network sites and photo sharing portals upload millions of images per day. In many cases, this leads to serious privacy threats. The images reveal not only the personal relationships and attitudes of the user who uploads the images, but those of other persons displayed in the images as well. In this paper, we propose a system architecture for privacy-aware image sharing. Our approach is based on the ImageNotion application, which combines automated processes to create high-quality semantic image annotations.

The dynamics of complex social networks has become one of the research areas of growing importance. The knowledge about temporal changes of the network topology and characteristics is crucial in networked communication systems in which accurate predictions are important.The local network topology can be described by the means of network motifs which are small subgraphs – usually containing from 3 to 7 nodes. They were shown to be useful for creating profiles that reveal several properties of the network. In this paper, the time-varying characteristics of social networks, such as the number of nodes and edges as well as clustering coefficients and different centrality measures are investigated. At the same time, the analysis of three-node motifs (triads) was used to track the temporal changes in the structure of a large social network derived from e-mail communication between university employees. We have shown that temporal changes in local connection patterns of the social network are indeed correlated with the changes in the clustering coefficient as well as various centrality measures values and are detectable by means of motifs analysis. Together with robust sampling network motifs can provide an appealing way to monitor and assess temporal changes in large social networks.

A number of studies on ontology editing tools and ontology-based applications have been proposed for automatically processing knowledge and information. However, the existing methodologies and tools for dealing with ontologies have assumed that the system is restricted to a single user. Main motivation of this paper is to foster collaborations between users, because ontology building is an expensive task. Thereby, in this paper, we present a web-based ontology construction and integration system, which is called OntoCS, to support collaborative interactions between people during creating ontologies. Particularly, inexpert users can collect available language resources from the web to describe concepts in a (even unfamiliar) domain. We believe that this collaborative process is implementing collective intelligence. In conclusion, we have shown that the proposed OntoCS system can efficiently edit and manage multiple ontologies over time.

The web ontology language (OWL) has become a W3C recommendation to publish and share ontologies on the semantic web. In order to infer implicit information (classification, satisfiability and realization) of OWL ontology, a number of OWL reasoners have been introduced. Ontology classification is to compute a partial ordering or hierarchy of named concepts in the ontology using the subsumption testing. Most of the reasoners use both top-down and bottom-up searches using subsumption testing for ontology classification. As subsumption testing is costly, it is important to ensure that the classification process uses the smallest number of tests. In this paper, we propose an enhanced method of optimizing the ontology classification process of ontology reasoning. Our work focuses on two key aspects: The first and foremost, we describe classical methods for ontology classification. Next, we present description of the enhanced method of optimizing the ontology classification and the detailed algorithm. We evaluate the effect of the enhanced method on four different types of test ontology. The enhanced search method shows 30

The analysis of error propagation is of fundamental importance to assure safe operation and management of abnormal situations in any distributed information system. In this paper, the quantitative and qualitative methods are proposed to analyze possible error propagation scenarios based on different topologies, error types and probability distributions. The most interesting from our point of view is the course of error propagation in simple structures that are contained in more complex ones. These complex structures, which have attracted the attention of scientists for many decades, are traditionally analyzed with the use of formalisms from graph theory. Certain types of graphs are often used to model naturally occurring complex structures, such as social networks. Graph-theoretic approach proved successful when applied to social networks and other naturally occurring complex networks. The research was verified based on the experiments conducted on simulation model. The results provide some ideas of robustness – the knowledge how to design the most error resistant architectures in complex environments.

This article discusses the challenge of the Integrated Project Support Environment development. Such an environment could be successfully applicable to a variety of software development project types. Since there are many software development approaches available it is important to choose and use an appropriate development methodology, methods and tools for a given project. The proposed Integrated Project Support Environment allows developers to configure software technology using formal specification to be used for project development.

With the Internet facing the growing problem of information overload, the large volumes, weak structure and noisiness of Web data make it amenable to the application of machine learning techniques. After providing an overview of several topics in text categorization, including document representation, feature selection, and a choice of classifiers, the paper presents experimental results concerning the performance and effects of different transformations of the bag-of-words document representation and feature selection, on texts extracted from the dmoz Open Directory of Web pages. Finally, the paper describes the primary motivation for the experiments: a new meta-search engine CatS which utilizes text categorization to enhance the presentation of search results obtained from a major Web search engine.

Assembler Encoding is Artificial Neural Network encoding method. To date, Assembler Encoding has been tested in two problems, i.e. in an optimization problem in which a solution is in the form of a matrix and in the so-called predator-prey problem in which the task of ANN is to control agent-predators whose common goal is to capture a fast moving agent-prey. The next problem in which Assembler Encoding was tested is the inverted pendulum problem. In the experiments Assembler Encoding was compared to several evolutionary and reinforcement learning methods. The results of the tests are presented at the end of the paper.