VOLUME 25 NUMBER 3 2001

Abstracts:

Object-Oriented frameworks provide a powerful technique for developing groups of related applications. Individual designers may develop applications tailored to their specific needs by providing appropriate definitions for the {\em hook} methods, while reusing the design built into the framework. One important problem in using this approach is the lack of suitable techniques for precisely documenting the behavior of such frameworks. If such techniques were available and were used to document framework behavior, application developers would be able to more precisely understand the framework, be able to more effectively build their applications from the framework, and be able to reason more reliably about the behavior of these applications. Our goal is to develop such techniques. The techniques we develop will allow the framework designer to specify the framework behavior precisely, and allow the application developer to `plug-in' information about the behavior of the hook methods into the specification of the framework to arrive at the behavior of the application. We illustrate the technique by applying it to a simple case study. (pp. 297-308)

Keywords: Object-oriented application frameworks, software reuse, formal specifications, verification

Design patterns are powerful design and reuse tools in software development. However current usage of patterns seem to employ a small portion of the enclosed potential. Commonly, pattern usage is limited to the manual customization of just a few legacy patterns, or documentation of existing designs. We think that a more effective use of the patterns requires standardization and formalization of the pattern utilization. In this work, we propose a formalism based on a fully generalized and purely symbolic interpretation of patterns, which focuses on the representation of reusable knowledge. (pp. 309-317)

Keywords: design patterns, software reuse, components

An original approach to support of knowledge management within an organization is presented in this article. This approach has been designed, implemented in form of the KnowWeb system and tested on various pilot applications. Special attention is paid to organizations with distributed environment. For this purpose an experimental system for support of mobile agents that combines the power of high-level distributed programming with the mobile agent paradigm has been proposed and is presented here as well. Finally, experiences from two of the KnowWeb pilot applications as well as further application possibilities in the area of e-Democracy are sketched. (pp. 319-328)

Keywords: knowledge management, knowledge modelling, mobile agents, agent collaboration, e-Democracy

Recently, several search engines have been established to help people find interesting information among the rapidly increasing number of web pages over the Internet. To obtain useful and reasonable searching results, users may submit queries with more than one query terms combined by a Boolean expression, supported by all existing search engines. However, these search engines all put the same emphasis on each query term combined by the Boolean expression. That is, for the identical queries, different users would obtain the same searching results. This contradicts the fact that different users usually have different searching interests even with the same queries. In other words, a useful search engine nowadays should allow users to emphasize each query term unequally to get the more reasonable and individual searching results. In this paper we propose an efficient approach, named Extreme Score Analysis method (ESA method), to solve this problem. ESA method uses web pages' original scores to derive users' top K web pages when each query term is assigned with a different weight. Moreover, we improve ESA method and further propose Extreme Score Analysis Inverse method (ESIA method), which can efficiently find users' top K interesting target ranks when these users assign different weights to each query term (pp.329-340)

Keywords: search engine, term weighting function, similarity function; ranking; Boolean expression

A tested recipe for executing academic Classificatory Data Mining challenges is described. The recipe includes a way of assessing the quality of all submissions using ROC curves. The challenge methodology is focused towards selecting the best \emph{techniques} for the problem domain offered in the challenge. Modifications to the recipe so that the focus is on obtaining the best \emph{model(s)} are discussed. The modified scheme could be used as the basis for a form of ``collaborative'' data mining. (pp. 343-347)

Keywords: Data Mining, Challenge Data Mining, ROC Analyses, Optimal Models, Near-optimal Models

The advent of Internet Banking has shown the importance of effective method of authenticating a users in a remote environment. There are many different countenances to contemplate when examining Internet based security. One of the most tried and trusted techniques of protecting the safety of systems and data is to control people's access. The foundation for such measures is authentication. Specifically for Internet banking there is a real need for a way to uniquely identify and authenticate users without the possibility of their authenticity being cloned. This paper proposes a framework concerning how to identify security requirements for Internet Banking. (pp. 349-355)

Keywords: E-Commerce, Computer Security, Internet Banking, Authentication

Media space consists from many different factors fighting for the attention of customer population in a certain environment. Common problem in bigger environments (or countries) is that datasets describing complete media space is hard or almost impossible to get since the detailed picture is too complex or to expensive to compose. However, this is not the case in environments, which are smaller, and is therefore easier to collect the data. We have access to the data entirely describing the media space of population of 2 million people. Because of the language and economy this media space functioning relatively independently from different factors, specially outside the country. The data was collected by Media Research Institute Mediana. The database consists of 8000 questionnaires, gathered in 1998. The sample and the questionnaires were made by comparable research standards. In this paper we will discuss different type of questions, which might become in a great assistants in unfolding the media groups, audience fluctuation and profound understanding of happenings in media space, as well as for their predictions (pp. 357-363)

Keywords: data mining, media space data, data analysis

This paper describes the first steps in analysis of millions of baskets collected over the past year from a retail grocery chain containing hundreds of stores. Each record in the data set represents an individual item processed by an individual checkout laser scanner at a particular store at a particular time on a particular day. In order to get some insights in the data, we used several different approaches including some statistical analysis, some machine learning, and some data mining methods. The sheer size of the data set has forced us to go beyond usual data mining methods and utilize {\em Meta-Mining}: the post processing of the results of basic analysis methods. (pp. 365-369)

Keywords: Data Mining, Meta Mining, market basket analysis, association rules, decision trees

Electronics, engineering and medicine expert collaboration as rule requires knowledge of both the remaining fields from each of experts to result in a useful application. Ergonomics and work process management are the goals achieved by such approach. Comprehensive analysis of work process, influenced by environmental factors and output, measured as a function of these influences, is a cornerstone of these multimedia CD compilation, presented both in a textbook and handbook format. In the electronic publication study contents are combined with practical experience of efficient Slovenian companies, which were prepare to present a piece of their experience.(pp. 371-374)

Keywords: electronic publication, multimedia, education, ergonomics

The paper discuses the role of the Organized Anarchy paradigm and the Garbage Can Model in strategic decision-making, and extends the idea of Organized Anarchy to the Informed Anarchy with unclear technology of allocation and dissemination of information, incomplete understanding of information, and fluid participation of information in decision processes. An example of the computer simulation is briefly presented, but the discussion in this paper is limited to the relation between the level of organization anarchy, load of problems, formal and informal information systems, and efficiency of decision-making. The models suggest that managers could enrich their decision-making by making their organization function like a net in which they catch "ingredients" needed for strategic planning and efficient decision-makings. Such nets are intelligently employed and motivated members of the organization. (pp. 375-379)

Keywords: organized anarchy, informed anarchy, strategic management. Petri net, computer simulation

This paper analyses main legal problems that are related to formation of e-contracts. Because of the different legal consequences it is highly important for parties to be aware and understand the basic phases of contract formation process. For a contract to exist, usually one party must have made an offer, and the other must have accepted it. Once acceptance takes effect, a contract will usually be binding on both parties. A variety of procedures are available for forming an electronic contract, such as e-mail, EDI etc. Therefore, we have established that an electronic communication could create a legally binding contract. Business and other contracting parties should define and articulate most of the rules that will govern electronic commerce. Internationally, UNCITRAL Model law establishes rules and norms that validate and recognize contracts formed through electronic means, sets default rules for contract formation and governance of electronic contract performance. (pp. 381-386)

Keywords: e-commerce, e-contract, offer, acceptance, formation of contract

TIn this paper we present and discuss the state of intelligent system applications, with special emphasis on Slovenia. We take the viewpoint that intelligent systems facilitate a new qualitative step towards the information society. Thus, they provide an important opportunity for any country to intensify its information, economic and social development. We present some intelligent system applications developed or being developed in our group: (1) EMA, an employment agent, (2) GIVE, a semantic web speaking system, and (3) DEXi, a decision modeling system. Based on past experience, we highlight the principal problems that occur in the development and exploitation of intelligent systems, and suggest improvements for the future. (pp. 387-392)

Keywords: intelligent systems, intelligent agents, applications, information society

Finding the optimal level of abstraction at which to document the architecture of a system has long been a problem in software engineering, particularly for large and complex systems. In this paper we argue that providing just a single abstraction level is inappropriate, and that instead, multiple architectural descriptions should be developed and documented, each capturing a specific aspect of a system's realization at a particular level of abstraction. Further, we argue that such a stratified architecture is especially valuable when used to organize a framework. After explaining the basic motivation for the work, and defining the basic principle of stratification, the paper illustrates the approach in conjunction with a small case study. The paper then discusses the methodological issues associated with the creation, application and maintenance of stratified frameworks. (pp. 393-401)

Keywords: components, refinement, architecture, stratification

Research results in machine learning have confirmed the possibility of reconstruction of human skill, resulting in similar or even improved performance of the same skill by a computer program. The methods that build symbolic models of skills are preferred if generated models are to be used to help humans in improving their own skill. The paper discusses these possibilities with the objective of accelerating and enhancing the process of human skill development. Experiments in learning control of dynamic systems are presented and wider potential of applications in assisting learning of skill is discussed. (pp. 403-408)

Keywords: human learning, modelling of human skill, machine learning, transfer of human skill, dynamic systems control

Probabilistic reasoning with belief (Bayesian) networks is based on conditional probability matrices. Thus it suffers from NP-hard implementations. In particular, the amount of probabilistic information necessary for the computations is often overwhelming. So, compressing the conditional probability table is one of the most important issues faced by the probabilistic reasoning community. Santos suggested an approach (called linear potential functions) for compressing the information from a combinatorial amount to roughly linear in the number of random variable assignments. However, much of the information in Bayesian networks, in which there are no linear potential functions, would be fitted by polynomial approximating functions rather than by reluctantly linear functions. For this reason, we construct a polynomial method to compress the conditional probability table in this paper. We evaluated the proposed technique, and our experimental results demonstrate that the approach is efficient and promising. (pp. 409-419)

Keywords: Probabilistic Reasoning; Belief Network; Fuzzy Reasoning; Compressibility of Information; Encode Technology

A meta-heuristics for escaping from local optima to solve constraint satisfaction problems is proposed, which enables self-adaptive dynamic control of the temperature to adjust the locality of stochastic search. In our method, several groups with different temperatures are prepared. To each group the same number of candidate solutions are initially allotted. Then, the main process is repeated until the procedure comes to a certain convergence. The main process is composed of two phases: stochastic searching and population tuning. As for the latter phase, after evaluating the adaptation value of every group, migration of some number of candidate solutions in groups with lower values to groups with higher values are induced. Population migration is a kind of parallel version of simulated annealing, where several temperatures are spatially distributed. Some experiments are performed to verify the efficiency of the method applied to constraint satisfaction problems. It is also demonstrated that population migration is exceptionally effective in the critical region where phase transitions occur. (pp. 421-429)

Keywords: constraint satisfaction, search algorithms, stochastic search, meta-heuristics

The introduction of up-to-date information technologies in all aspects of business activities, maritime industry included, represents the basic prerequisite of success. Due to extremely difficult situation in maritime industry in most under-developed countries, there are no adequate information systems. Wishing to present the current situation and the possibilities of application of up-to-date scientific methods in projecting and implementing information systems, this paper deals with: (a) Up-to-date methodological approach to projecting information systems through CSF (Critical success factor), E-M (End- Means), BSP (Business System Planning) methods and their comparative analysis; (b) Present condition of information systems in shipping companies of most under-developed countries; (c) Possibilities of modeling the existing systems through determining the business goals and projecting matrix relations between business organizational units, processes, and data classes; (d) Upgrading of these relations by HIPO (Hierarchy Input Process Output) method, referring to input-output diagrams of certain business processes; (e) Possible architecture of integrated shipping company information systems; (f) Review of integrated ship information systems as a key subsystem of a shipping company. (pp. 431-438)

Keywords: information system, modeling, shipping company