Mieczysław L. Owoc

Validation and verification of knowledge-based systems: report on EUROVAV99

Knowledge-Based (KB) technology is being applied to complex problem solving and safety and business critical tasks in many application domains. Concerns have naturally arisen as to the dependability of Knowledge-Based Systems (KBS). As with any software, attention to quality and safety must be paid throughout development of a KBS, and rigorous Verification and Validation (V&V) techniques must be employed. Research in V&V of KBSs has emerged as a distinct field only in the last decade, and is intended to address issues associated with quality and safety aspects of KBSs, and to provide such applications with the same degree of dependability as conventional applications. In recent years, V&V of KBSs has been the topic of annual workshops associated with the main AI conferences, such as AAAI, IJCAI and ECAI.

Knowledgebases: A Management Context and Development Determinants

Contemporary computers give huge possibilities for processing different structures of data e.g. multimedia data types and very unconstrained in construction - object types. Moreover these different structures are processed using methods called intelligent. Knowledgebases (a specific sort of intelligent systems), as a consequence of variety of solutions, belong ambiguous terms. Principally, applications with a stored knowledgebase and a reasoning mechanism - to stress its structural components – are termed knowledgebases. Directions of the research concerning knowledgebases and expressing their peculiar characteristics (coming from the main message of scientific formation in the past) are presented in this paper. The starting point of the paper is observed impact of management processes on intelligent systems and vice versa. Managerial context (including classical, behavioural, quantitative, integrating contemporary and future management approaches) of knowledgebases became the ground to draft development of applications of this sort framework. Taking into account research on knowledgebases, a set of knowledgebase development determinants is discussed.

Validation of Rule-Based Systems Generated by Classification Algorithms

Validation of a knowledge base is an important aspect of the knowledge-based systems (KBS) development procedure, which aims to assure the system’s ability to reach correct results. Very promising seems to be knowledge-bases (KBs) generated by classification algorithms as an example of machine-learning approach. The paper addresses the issues of verification and evaluation of KBs existing as rule-based systems and generated with classification algorithms. The framework consists of three steps. The first one relies on creation a set of rules using two tested algorithms: C4.5 and GIMS and then transformation rules into decision tables. In the second one, rules are verified, taking into account two criteria: completeness and consistency. Finally, during the last step, the set of rules is evaluated using two additional criteria: adequacy and reliability. The classification problem refers to bank customers, which apply to get credit; their applications can be approved or refused. Certain unique features of generated rules are shortly commented in a summarisation.

On Principles of Knowledge Validation

Validation is a critical process in the whole knowledge-based system life cycle. A knowledge base incorporated into such systems has to be verified or (more generally) validated. There have been many approaches to develop specialised procedures and techniques, aimed at assuring the highest level of knowledge quality. Keeping in mind “knowledge validation mappings”, we believe a more global view is necessary to facilitate applying the proper techniques, so the paper deals with practical guidelines of knowledge validation (KV). Facing the most popular techniques of KV: decision tables-based, decision trees-based or nets-based with the criteria set to be utilised, we try to define certain principles useful in the validation procedures referring to two levels: general and detailed. The first one refers to paradigms, which arise from interrelationships among the crucial components of the KV process (procedures, approaches and criteria). The detailed principles are addressed to specific forms used for knowledge representations: rules, frames, neural nets and others.

Knowledge Granularity and Representation of Knowledge: Towards Knowledge Grid

Knowledge granularity, usually identified with the size of knowledge granules, seems to be real challenge for knowledge consumers as well as for knowledge creators. In this paper, relationships between knowledge granularity as a result of different ways of a knowledge representation are considered. The paper deals with the problem of developing knowledge grid in the context of encapsulation of knowledge including different dimensions and measures. The origin of the problem is discussed in the first section stressing flexibility of knowledge interpretations. Concepts of knowledge granularity (also from formal point of view) are presented in the next section. The nature of well represented knowledge is considered in the next chapter with references to granularity of knowledge. In the last part of the paper the question of formulating knowledge granularity in the context of knowledge grid concepts is discussed. This document comprising guidelines for authors is divided into several sections.

Knowledge portal for exclusion process services

Exclusion phenomenon in common understanding denotes processes in which some group of people or individuals are permanently blocked from resources (mostly considered as social exclusion). For sure such sort of phenomena is observed as unwanted not only from “outsiders” but also from local and global society point of view. The exclusion (and its antonym inclusion) phenomenon can be investigated including many aspects: starting from identifying exclusion as a process, multidimensional research in this area up to solutions available in the domain. In order to be successful in overtaking this phenomenon groups and institution involved in this process should be supported by ICT solutions. The paper consists of five parts which gradually present context of the problem and proposed solutions. After short introduction concerning research background the discussed concept of exclusion processes is presented. In the main section real examples of exclusion processes are investigated which allows to define assumptions of knowledge portal architecture and some examples of performed services. It creates opportunities for formulation final conclusions about the necessity and usability the developed platform.

Validation model for discovered web user navigation patterns

Information society with huge number of everyday acting participants becomes valuable source of surfer behavior research. Especially web server logs can be used for discovering knowledge useful in different areas. Knowledge acquired from web server logs in order to generate solutions has to be validated. The aim of this paper is presentation of web usage mining as a rather complex process and in this context elaboration of validation model. On a basis of iterative and hybrid approach to discover user navigation patterns the concept of generated knowledge validation model is proposed. Some experiments on real website allow to define a new method of generated association rules refinement including specific knowledge validation techniques. Some constraints as discovered knowledge validation criteria are defined.

Contextual Knowledge Granularity

Knowledge granularity is often regarded as one of the essential factors of knowledge repositories basically in terms of ways of knowledge gathering and storing as well as its usability. The aim of this paper is to discuss the importance of this phenomena in the case of contextual classification. This kind of directed granulation by context gives possibility to generate new and intelligent knowledge structure, to see a problem through many context simultaneously perspectives or through one context perspective. A part of the investigation of usability of context-based approaches in creation knowledge structures interrelationships between knowledge granularity and effectiveness of classification tasks is discussed.

Design of "Business Informatics" Study Program Model, Curriculum and Perspectives

In this paper a study program of “Business Informatics” at the Wroclaw University of Economics, Poland, is presented. The structure of subjects taught and organization of the curriculum for Bachelor and Master levels are detailed. The education model is presented along with descriptions of profession-oriented modules. The authors discussed also a comparison of the new program of Business Informatics and a previous one, called “Informatics and Econometrics” that was common and official program of applied informatics in business schools in Poland. The perspectives and the constraints of program implementation of the new field of study are indicated.

Towards better understanding of context-aware knowledge transformation

Considering different aspects of knowledge functioning, context is poorly understood in spite of intuitively identifying this concept with environmental recognition. For dynamic knowledge, context especially seems to be an essential factor of change. Investigation on the impact of context on knowledge dynamics or more generally on the relationship between knowledge and its contextual interpretation is important in order to understand knowledge dynamics. The aim of this paper is to research and examine the nature of knowledge transformation (a specific sort of life-cycle), and to identify contextual factors affecting knowledge dynamics.