Roman Simiński

Extraction of Rules Dependencies for Optimization of Backward Inference Algorithm

This work presents the modification of backward inference algorithm for rule knowledge bases. Proposed algorithm extracts information of internal rules dependencies and performs only promising recursive calls. Optimization relies on reducing the number of rules searched for each run of inference and reducing the number of unnecessary recursive calls. We assume that the rule knowledge base itself contains enough information, which allow to improve the efficiency of the classic algorithms of the inference and we propose the decision units conception as tool for extracting and modeling such information. The first part of the work briefly presents backward inference algorithms in its classical version, next part of the work describes the decision units conception, then the utilization of decision units in optimization of inference algorithm is described and the modified versions of algorithm are presented. The preliminary evaluation of modified versions of algorithm finish presented work.

Towards a practical approach to discover internal dependencies in rule-based knowledge bases

In this paper, we intend to introduce the conception of discovering the knowledge about rules saved in large rule-based knowledge bases, both generated automatically and acquired from human experts in the classical way. This paper presents a preliminary study of a new project in which we are going to join the two approaches: the hierarchical decomposition of large rule bases using cluster analysis and the decision units conception. Our goal is to discover useful, potentially implicit and directly unreadable information from large rule sets.

Knowledge Mining Approach for Optimization of Inference Processes in Rule Knowledge Bases

The main aim of the article is to present modifications of inference algorithms based on information extracted from large sets of rules. The conception of cluster analysis and decision units will be used for discovering knowledge from such data.

Decision Units as a Tool for Rule Base Modeling and Verification

Expert systems are problem solvers for specialized domains of competence in which effective problem solving normally requires human expertise. The transition of expert systems technology from research laboratories to software development centers highlighted the fact that the quality assurance for expert system is a very important issue for most real-word problems. Although the basic verification concepts are shared by software engineering and knowledge engineering, verification methods of conventional software are not directly applicable to expert systems and the new, specific methods of verification are required. The main aim of this work is to present our own rule base verification method. In our opinion the decision units conception allows us to consider different verification and validation issues together. Thanks to properties of the decision units we can perform different verification and validation actions during knowledge base development and realization.

Verification of Rule Knowledge Bases Using Decision Units

Expert systems are being developed and used in a wide variety of disciplines throughout the world. Much attention has been paid to check the reliability of expert systems. Verification of knowledge bases has emerged as a significant problem in the development of expert systems. We argue that knowledge bases verification can not be delayed until the final knowledge base realization and we suppose that verification should be performed incrementally and should be included into the development process. In this paper we introduce the main assumptions of decision unit conception and we briefly describe how we are going to apply decision units in knowledge bases verification and validation.

Multivariate Approach to Modularization of the Rule Knowledge Bases

This article introduces the multivariate approach to modularization of the rule knowledge bases. The main difference between proposed approach and other known modularization methods consists in the fact that proposed idea allow us to use different modularization strategies, according to the current requirements specified by the expert or knowledge engineer. This work describes the formal modularization model, an example of rule base modularization variants and short comparison with existing modularization methods.

Goal-Driven Inference for Web Knowledge Based System

Traditional knowledge based systems were developed as the desktop applications. Meanwhile, web applications have grown rapidly and have had significant impact on the application of such systems. In the presented work, we introduce the modified goal-driven inference algorithm which allow us to divide some parts of them into the client and server layers of the web application. Proposed approach assumes that the rule knowledge base is decomposed into the decision oriented group of rules. We argue that the knowledge base in the form of such rules group contains enough information, which allows to divide inference into the client and server side, ensuring the convenience and the effectiveness.

Application of Decision Units in Knowledge Engineering

In this paper we shall present the decision units idea that, allow us to divide a set of rules into subsets – decision units. This paper gives attention to decision units in chosen problems of knowledge engineering. We present our own rule base verification method based on decision unit conception. The decision units are simple and intuitive models describing relations in rule knowledge base, being direct superstructure of a rule-base model. The unit’s usage offers the knowledge engineer’s support on the technical design and base realization level.

KBExplorator and KBExpertLib as the Tools for Building Medical Decision Support Systems

Medicine had been considered a good domain in which the concepts of rule-based decision support system could be applied. The early medical decision support systems were designed over forty years ago. Since that time, many different methods were proposed in the decision support area. Regardless of the development of different non knowledge-based methods, the rule representation and inference on the rules bases are still popular. In this paper the KBExplorator system and KBExpertLib software library are introduced in the context of medical decision support system implementation. The KBExplorator system may be considered as tool for building medical knowledge bases. This system is designed for knowledge engineers and domain experts, which are responsible for creating the knowledge base for particular problem. The software library KBExpertLib is a tool for programmers to develop software which utilize the knowledge bases designed with use of KBExplorator. The main properties and methods of practical usage of KBExplorator and KBExpertLib are described as well as experiments focused on the software effectiveness evaluation.

Physical Knowledge Base Representation for Web Expert System Shell

Web applications have developed rapidly and have had a significant impact on the application of systems in many domains. The migration of information systems from classic desktop software to web applications can be seen as a permanent trend. This trend also applies to the knowledge based systems. This work is a part of the KBExplorator project – the main goal of this project is to provide a complete and easy to use web-based tool for the development of expert systems. The evaluation of the rules searching effectiveness in the proposed physical rule base model is the first experimental aim of this work. Experiments will be conducted to determine the duration of retrieving a single rule or group of rules in large rules sets. Decomposition of the rule knowledge base into the relational database is also a crucial issue of this work and therefore the presentation of the data model is the second goal of this work. The usage of a relational database in the web-based application is obvious, but its usage as the physical storage for the rule base is described in relatively small number of publications. Proposed decomposition conception and the model presented in this work has not been previously described. The positive results of experiments presented in this work allow us to continue the development of the system – in the next revision, the database interface layer will be implemented with the usage of a specialized API. This proposed software architecture allow us to transparently change the database engine as well as the programming language currently used in the application layer of the system.