Abstracts:

Lars Willmes and Peter Krause
NuTech Solutions GmbH
Martin-Schmeisser-Weg 15
44227 Dortmund, Germany
E-mail: willmes@nutechsolutions.de, krause@nutechsolutions.de

The basic variants of evolution strategies, a special instance of evolutionary algorithms, are discussed in this paper. Gleaned from the model of organic evolution, evolution strategies are characterized by the additional self-adaptive process that fine-tunes their strategy parameters during optimization. This property is a fundamental ingredient for the application to challenging engineering applications involving resource-intensive simulation runs. For one instance of such applications, the single-criterion and multi-criterion airfoil design problem, the results of an evolution strategy are presented and discussed in this paper. (pp. 337-344)

Keywords: Evolutionary algorithm, multi-criterion optimization, airfoil design

Population based evolutionary algorithms (EA) are frequently used to optimize on multimodal functions. A common assumption is that during search several subpopulations might coexist in different attraction regions of the search space. Practical experience and takeover--time considerations suggest that this is not true in general. We therefore analyze the stability of subpopulations within a simplified EA on a two-attractor model, focusing on two extreme cases: (1) Function values of both local minima are exactly the same and (2) function values on the first attractor are always better than on the second. Realistic scenarios for bimodal optimization are assumed to be located in between these two extremes, such that upper and lower bounds for extinction times can be estimated, e.g. by Markov chain analysis and empirical studies. The obtained results provide new insights into the effect of μ+,λ selection on the stability of subpopulations and the effect of genetic drift. Moreover, the effect of idealized niching on the same scenarios is investigated, leading to an immense increase of the EA's ability to perform concurrent search. Our model and the findings based thereupon do not depend on the number of problem dimensions. (pp. 345-351)

Keywords: Evolutionary algorithms, multimodal fitness landscape, niching techniques, random genetic drift

We present an evolutionary algorithm approach to schedule optimization for a group of production lines in a car factory. Schedules are evaluated with respect to the energy consumption over peak demand periods, while the task is to minimize the energy costs by appropriately scheduling the interruptions of processes on the lines. Tests on real problem instances show this approach gives near-optimal schedules in acceptable time. (pp. 353-357)

Keywords: evolutionary algorithm, production scheduling, energy consumption, peak energy demand, cost minimization, automobile industry

In this paper we present an evolutionary algorithm for solving the nutrition problem of composing and balancing healthy meals. We treat this problem as a single-objective and multiconstrained fractional knapsack problem that is easy to formulate, yet, its decision problem is in the class of NP-complete problems. In other words, some heuristic algorithm is required to provide good problem solutions in reasonable (polynomial) computational time. We applied a genetic algorithm and modified its parameters to yield high-quality and reliable solutions (healthy and balanced meals) that respect multiple weakly-correlated dietary recommendations and guidelines and include as much seasonal functional foods as possible. Functional foods contain physiologically active compounds that provide health benefits beyond their nutrient contributions. (pp. 359-364)

Keywords: Optimization, multiconstrained fractional knapsack problem, genetic algorithms.

Jaro Berce
RS Government Office for EU Affairs,
Šubičeva 11
SI - 1000 Ljubljana, Slovenia
E-mail:jaro.berce@gov.si

A research work presented in the paper establishes that the overall goal in public sector is better governance - helping to narrow the knowledge divide and its consequences between societies and also within each society so to stipulate eDemocracy. The learning process narrow divides between those societies who can accumulate, manage, when/how to share, and when to use knowledge and those who face obstacle and challenges at one ore more stages in the sequence from acquiring knowledge to use. A strong linking of all factors (social, economic, technology) and their proper transform order are the winning solutions. The answer lies with recognition of peoples dimension: knowledge (people's and organization's), sharing and networking knowledge and information, learning, and organizational issues encouraged with economic dimension (rewarding, budget, and financing) and driven by new technology solutions. Therefore, Information and Communication Technology is a general "driving" force that makes Knowledge Management, Learning Organization, and eGovernance more important activity than in the past. (pp. 365-370)

Keywords: iGovernment, eGovernment, knowledge management, learning organization, information communication technology, eGovernance, public administration

Žiga Mahkovec
Faculty of Computer and Information Science
Tržaska cesta 25
SI-1001 Ljubljana, Slovenia
E-mail:ziga.mahkovec@klika.si

We present a software agent for categorizing and geolocating news articles. The articles are retrieved from different on-line news sources, such as Google News, Reuters and BBC News. They are parsed, categorized based on crime threat, geographically located and rendered in an SVG widget. The agent is implemented in Java, using the Scalable Vector Graphics markup language to render the user interface. Text categorization is performed using the Support Vector Machine (SVM) method, with test data from the Reuters RCV1 corpus. The GEONet Names Server and Digital Chart of the World databases are used to geolocate the news articles. (pp. 371-374)

Keywords: Text categorization, support vector machine, RCV1, geolocation, GIS, SVG

Marko Grobelnik and Dunja Mladenić
Jožef Stefan Institute
Jamova 39
SI-1000 Ljubljana, Slovenia
E-mail: marko.grobelnik@ijs.si, dunja.mladenic@ijs.si

This paper presents a system for visualization of large amounts of new stories. In the first phase, the new stories are preprocessed for the purpose of name-entity extraction. Next, a graph of relationships between the extracted name entities is created, where each name entity represents one vertex in the graph and two name entities are connected if they appear in the same document. The graph of entities is presented as a local neighborhood enriched with additional contextual information in the form of characteristic keywords and related name entities connected to the entity in the focus. Operations for browsing a graph are implemented to be efficient enabling quick capturing of large amounts of information present in the original text. (pp. 375-380)

Keywords: text visualization, context visualization, document clustering, large collection of news stories

Jure Ferlež and Matjaž Gams
Jožef Stefan Institute
Department of Intelligent Systems
Jamova 39
SI-1000 Ljubljana, Slovenia
E-mail: jure.ferlez@ijs.si, matjaz.gams@ijs.si

This paper analyses a measure of semantic relatedness between two words. Our measure is based on shortest path between synsets in WordNet v2.0. It uses all available links in WordNet v2.0 and is implemented by a bidirectional breadth-first algorithm. Experimental evaluation and comparison with a benchmark set of human similarity judgments demonstrates that the simple measure applied on WordNet v2.0 performs better than the more complicated approaches mostly combining an IS-A taxonomy with the notion of shared information content extracted from corpora. One explanation is that our pretty basic method efficiently exploits all the available links in WordNet v2.0, while other measures, although more complicated and advanced, do not make good use of the new derivational links added to WordNet in the latest version 2.0. (pp. 381-386)

Keywords: semantic relatedness, semantic distance, WordNet

Marko Bohanec, Sašo Džeroski and Martin Žnidaršič
Jamova 39
SI-1000 Ljubljana, Slovenia
E-mail: marko.bohanec@ijs.si, saso.dzeroski@ijs.si

Antoine Messéan
CETIOM, Centre technique interprofessionnel des Oléagineux métropolitains and INRA Eco-Innov
POB BP1, Centre de Grignon
FR-78850 Thiverval, Grignon, France
E-mail: messean@cetiom.fr

Sara Scatasta and Justus Wesseler
Wageningen University , Environmental Economics and Natural Resources Group
De Leeuwenborch, Hollandseweg 1
NL-6706 KN Wageningen, The Netherlands
E-mail: sara.scatasta@wur.nl, justus.wesseler@wur.nl

Modelling of economic and ecological impacts of genetically modified crops is a demanding task. We present some preliminary attempts made for the purpose of the ECOGEN project "Soil ecological and economic evaluation of genetically modified crops". One of the goals of the project is to develop a computer-based decision support system for the assessment of economic and ecological impacts of using genetically modified crops, with special emphasis on soil biology and ecology. The decision support system will be based on a rule-based model incorporating both economic and ecological criteria. In this paper we present some preliminary results of developing the integral model and describe four specific sub-models. The first two sub-models are concerned with ecology and assess the ecological impacts of various types of weed and pest control, respectively. The other two sub-models assess the economic impacts of cropping systems at the farm and regional level, respectively. All the models were developed using a qualitative multi-attribute modelling methodology, supported by the software tool DEX (pp. 387-392)

Keywords: qualitative multi-attribute modelling, genetically modified plants, cropping system, ecology, economy

Tomaž Kosar, Marjan Mernik and Viljem Žumer
University of Maribor
Faculty of Electrical Engineering and Computer Science
Slovenia
E-mail: tomaz.kosar@uni-mb.si, marjan.mernik@uni-mb.si, zumer@uni-mb.si

Pedro Rangel Henriques
University of Minho , Department of Computer Science, Portugal
E-mail:prh@di.uminho.pt

Maria João Varanda Pereira
Polytechnic Institute of Bragança, Portugal
E-mail:mjoao@ipb.pt

The paper presents a grammatical approach to software development. It supports formal software specification using attribute grammars, from which a rapid prototype can be generated, as well as the incremental software development. Domain concepts and relationships among them have to be identified from a problem statement and represented as a context-free grammar. The obtained context-free grammar describes the syntax of a domain-specific language whose semantics is the same as the functionality of the system under implementation. The semantics of this language is then described using attribute grammars from which a compiler is automatically generated. The execution of a particular program written in that domain-specific language corresponds to the execution of a prototype of the system on a particular use case. (pp. 393-405)

Keywords: software design and modelling, software development, context-free grammars, attribute grammars, rapid prototyping

Integration of information systems is a complex field where major challenges are semantic, process and technology related. Integration must be performed using methods, disciplines and activities that enable it to be effective in terms of costs and time - thus it should be supported by a well defined integration process. This article presents an information systems integration process model proposal with the goal to guarantee the quality of the integrated solution. The article focuses particularly on the integration specific disciplines: analysis of existing applications and integration design. (pp. 405-414)

Keywords: integration, process, EAI, B2B, information systems

Ralf Steinberger, Bruno Pouliquen and Camelia Ignat
European Commission - Joint Research Centre ( JRC )
Via E. Fermi, T.P . 267, 21020 Ispra ( VA ), Italy
E-mail: Ralf Steinberger@jrc.it, Bruno Pouliquen@jrc.it, Camelia Ignat @jrc.it

We are proposing a simple, but efficient approach for a number of multilingual and cross-lingual language technology applications that are not limited to the usual two or three languages, but that can be applied with relatively little effort to larger sets of languages. The approach consists of using existing multilingual linguistic resources such as thesauri, nomenclatures and gazetteers, as well as exploiting the existence of additional more or less language-independent text items such as dates, currency expressions, numbers, names and cognates. Mapping texts onto the multilingual resources and identifying word token links between texts in different languages are basic ingredients for applications such as cross-lingual document similarity calculation, multilingual clustering and categorisation, cross-lingual document retrieval, and tools to provide cross-lingual information access (pp. 415-423)

Keywords: multilingual text analysis; cross-lingual information access; information extraction; document similarity; categorisation; clustering; nomenclatures; thesauri; gazetteers; freely available resources

Melita Hajdinjak and France Mihelič
University of Ljubljana , Faculty of Electrical Engineering, Slovenia
Via E. Fermi, T.P . 267, 21020 Ispra ( VA ), Italy
E-mail: melita.hajdinjak@fe.uni-lj.si, france.mihelic@fe.uni-lj.si

Human-human and human-computer dialogues differ in such an important way that the data from human interaction becomes an unreliable source of information for some important aspects of designing natural-language dialogue systems. Therefore, we began the process of developing a natural-language, weather-information-providing dialogue system by conducting the {\em Wizard-of-Oz} (WOZ) experiment. In WOZ experiments subjects are told to interact with a computer system, though in fact they are not since the system is partly simulated by a human, the wizard. During the development of the weather-information-providing dialogue system this experiment was used twice. While the aim of the first WOZ experiment was, first of all, to gather human-computer data, the aim of the second WOZ experiment was to evaluate the newly-implemented dialogue-manager component. The evaluation was carried out using the PARADISE evaluation framework, which maintains that the system's primary objective is to maximize user satisfaction, and it derives a combined performance metric for a dialogue system as a weighted linear combination of task-success measures and dialogue costs. (pp. 425-429)

Keywords: natural-language dialogue systems, Wizard-of-Oz experiment, dialogue-manager evaluation, PARADISE evaluation framework

Cvetana Krstev
Faculty of Philology, University of Belgrade
Studentski trg 3, 11000 Begrade , Serbia and Montenegro
E-mail: cvetana@matf.bg.ac.yu

Duško Vitas
Faculty of Mathematics, University of Belgrade
Studentski trg 16, 11000 Begrade , Serbia and Montenegro
E-mail: vitas@matf.bg.ac.yu

Tomaž Erjavec
Department of Knowledge Technologies, Jožef Stefan Institute
Jamova 39, 1000 Ljubljana, Slovenia
E-mail: tomaz.erjavec@ijs.si

MULTEXT-East is a multilingual dataset for language engineering research and development. This standardised and linked set of resources covers a large number of mainly Central and Eastern European languages and includes the EAGLES-based morphosyntactic specifications, defining the features that describe word-level syntactic annotations; medium scale morphosyntactic lexica; and annotated parallel, comparable, and speech corpora. The most important component is the linguistically annotated corpus consisting of Orwell's novel ``1984'' in the English original and translations. MULTEXT-East has already seen several editions, with the latest one being Version 3, where the most important addition are the Serbian language resources, including the structurally annotated ``1984'', the morphosyntactic specifications, the morphosyntactic lexicon and the linguistically annotated ``1984''. The complete dataset, unique in terms of languages and the wealth of encoding, is extensively documented, and freely available for research purposes. (pp. 431-436)

Keywords: natural language processing, language resources, Serbian language, multilinguality

Franci Pivec
IZUM Maribor
E-mail: franci.pivec@izum.si

Vladislav Rajkovič
FOV Kranj
E-mail: vladislav.rajkovic@ijs.si

Andrej Jus
Infos Ljubljana
E-mail: andrej.jus@infos.si

The World Computer Congress (WCC) held in Ljubljana in 1971 played a very important role in the promotion of computer science. Slovenian school authorities enjoyed relative autonomy in the former Yugoslavia and this made it possible for them to promote computer education. Informal computer education and the introduction of computer science subjects into regular school curriculum started very early in the 1970s. Strong support from civil society was of significant importance in that process. This expressed orientation towards an information society was one of the major differences between Slovenia and the rest of the former Yugoslavia and one of the causes contributing to the attainment of Slovenia 's independence. (pp. 437-441)

Keywords: WCC; computer education; computer programming competition