Vladislav Rajkovič

Transforming arbitrary tables into logical form with TARTAR

The tremendous success of the World Wide Web is countervailed by efforts needed to search and find relevant information. For tabular structures embedded in HTML documents, typical keyword or link-analysis based search fails. The Semantic Web relies on annotating resources such as documents by means of ontologies and aims to overcome the bottleneck of finding relevant information. Turning the current Web into a Semantic Web requires automatic approaches for annotation since manual approaches will not scale in general. Most efforts have been devoted to automatic generation of ontologies from text, but with quite limited success. However, tabular structures require additional efforts, mainly because understanding of table contents requires the comprehension of the logical structure of the table on the one hand, as well as its semantic interpretation on the other. The focus of this paper is on the automatic transformation and generation of semantic (F-Logic) frames from table-like structures. The presented work consists of a methodology, an accompanying implementation (called TARTAR) and a thorough evaluation. It is based on a grounded cognitive table model which is stepwise instantiated by the methodology. A typical application scenario is the automatic population of ontologies to enable query answering over arbitrary tables (e.g. HTML tables).

Knowledge-based portfolio analysis for project evaluation

Abstract A computer-based expert system for the evaluation of research and development projects is presented. The system was developed for The Ministry of Science and Technology of the Republic of Slovenia and in the process of evaluation and selection of projects submitted to the annual competition for funds. The system is based on an adapted portfolio matrix that determines the position of each project with respect to its contents and feasibility. The aggregation of these criteria is carried out by a qualitative multi-attribute decision model that was developed using an expert system shell: DEX. The model consists of a tree of criteria, supplemented by if-then rules. In addition to describing these components, the paper presents and discusses a practical application of the system.

Context Sensitive Decision Support Systems

Textbook accounts and case studies of decision making in the social am organisational contexts of decision support typically concentrate on modelling the representation of the decision problem, and/or of infonnation which may be relevant in its solution and/or the motivations of interested parties, etc. They employ the discourses of the participants in the decision making, and of decision analysts and/or DSS designers, as accessories, useful in contextualising the representations which are the focus of the account This paper attempts to invert this process, focusing on the discourses employed in decision making, decision analysis and the design of decision support, frrst, identifying the kind of discourse employed in negotiating and constructing decision problem representations, at five qualitatively different levels in detennining prescriptions for action, and, then, investigating how these representations are employed as artefacts by those people who participate in the making of decisions and attempt to get them implemented in organisational contexts.

Evaluating options by combined qualitative and quantitative methods

Abstract This paper addresses two problems related to qualitative decision making: option ranking and non-sensitivity to small differences between options. In general, only a partial order of options can be established by a qualitative model, which might be insufficient particularly when the number of options is large. A qualitative model is also incapable of discriminating between slightly different options. In this paper, a solution is proposed that is based on an automatic construction of a quantitative evaluation model from the qualitative one. In addition to a qualitative class, a quantitative utility is obtained for each option, which is used to rank options within classes and to reflect the sensitivity to small differences between options.

Knowledge engineering techniques for utility identification

Abstract In this paper multiattribute decision making is discussed in terms of decision-making knowledge. Special emphasis is on identification (measurement) and verification of utility functions, and their use for evaluation of alternatives and explanation of evaluation results. Axiomatic and direct approach in utility theory are compared to the approach based on inductive learning techniques which are known from the field of artificial intelligence. Alternatives or their parts with the known utility are taken as learning examples in order to construct utility (function) knowledge. This approach is supported by a special expert system shell for utility knowledge modelling. It is implemented on a personal computer as a part of DECMAK system.

Five Decision Support Applications

This chapter presents five real-life applications of decision support methods and techniques, conducted within the SolEuNet Project. The problem areas were: (1) The selection of banks for the Slovenian National Housing Program, (2) Housing loan allocation, (3) diabetic foot risk assessment, (4) model development for the selection of information technology providers, and (5) the evaluation of research project proposals. The approach was based primarily on qualitative multi-attribute modeling, and was combined with databases and other general modeling methods. The bank selection case is presented in detail whereas the remaining four applications are presented quantitatively.

Evaluating Patients' Health Using a Hierarchical Multi-attribute Decision Model

Evaluation of a patients health status is an essential part of the healthcare process. For this purpose, Hendersons model of basic living activities (BLA) is often proposed as a set of criteria to be used in nursing. Despite its clarity and theoretical background, the model is only partially used in clinical practice. In this paper, we present the methodology for a hierarchical multi-attribute decision model to increase the practical efficiency of the BLA model. The result is a computerized model for the evaluation of a patients health status. This model was tested in clinical practice by 17 nurses in two health centres in Slovenia and a strengths, weaknesses, opportunities and threats (SWOT) analysis was carried out. The strengths included providing a holistic understanding of the nature and level of the nursing problems, enriching the documentation and reducing the possibility of overlooking something important. As a part of electronic documentation, this computerized model supports systematic patient data gathering and evaluation.

Evaluation models for housing loan allocation in the context of floats

In this paper, we analyze the impact of contextual changes to the size, structure and behavior of a qualitative multi-attribute model. The study involves an evaluation model for allocating housing loans in the Housing Fund of the Republic of Slovenia. So far, this model was used in 13 completed floats of loans for citizens. As a result of considerable contextual changes that occurred from float to float, the model had to be adapted accordingly. These changes were reflected in the modifications of model structure, used attributes, and decision rules at various levels of the model. Based on the analysis of time series of different model parameters, we articulate some guidelines for efficient model management, and justify the flexibility of the approach and its robustness to contextual changes. In particular, we show that a model can be designed so that relatively large changes in the decision environment cause minor and isolated changes in the model.

Qualitative Multi-attribute Modeling and its Application in Housing

Hierarchical multi-attribute models are aimed at the classification or evaluation of options that occur in a decision process. They decompose the problem into smaller and less complex subproblem and represent it by a hierarchy of attributes and utility functions. This article describes a qualitative modeling approach, which is supported by an expert system shell DEX and whose models are characterised by qualitative attributes and if-then rules. The approach is presented using a real-world case: allocation of housing loans at the Housing Fund of the Republic of Slovenia. The article focuses on the design, implementation and evolution of the corresponding multi-attribute models, and presents the lessons learned during the ten years of the system’s exploitation.

A Concept of Rule-Based Decision Support Systems

We present an outline of rule-based decision theory where decision knowledge is represented and handled as logical rules, with probability and/or fuzziness. The theory is based on the fundamental belief that people are able to express their opinion on preferences using rules. The paper presents some practical results on obtaining, representing, aggregating and verifying rule-based decision knowledge for a decision support system. We emphasise the differences between traditional decision theory and the rule-based theory.