Wyniki wyszukiwania - BazTech

1

Multi-agent blackboard architecture for supporting legal decision making

Szymański Ł., Śnieżyński B., Indurkhya B.

Computer Science

|

2018

|

Vol. 19 (4)

457–-477

EN

Our research objective is to design a system to support legal decision making using the multi-agent blackboard architecture. Agents represent experts that may apply various knowledge-processing algorithms and knowledge sources. Experts cooperate with each other using the blackboard to store facts about a current case. Knowledge is represented as a set of rules. The inference process is based on bottom-up control (forward chaining). The goal of our system is to find rationales for arguments that support different decisions for a given case by using precedents and statutory knowledge. Our system also uses top- -down knowledge from statutes and precedents to interactively query the user for additional facts when such facts could affect the judgment. The rationales for various judgments are presented to the user, who may choose the most appropriate one. We present two example scenarios in Polish traffic law to illustrate the features of our system. Based on these results, we argue that the blackboard architecture provides an effective approach to modeling situations where a multitude of possibly conflicting factors must be taken into account in the decision making.

2

The Heuristic Model Based on LPR in the Context of Material Conversion

Wilk-Kołodziejczyk D., Kluska-Nawarecka S., Nawarecki E., Śnieżyński B., Jaśkowiec K., Legień G.

Archives of Metallurgy and Materials

|

2017

|

Vol. 62, iss. 3

1603--1607

EN

High complexity of the physical and chemical processes occurring in liquid metal is the reason why it is so difficult, impossible even sometimes, to make analytical models of these phenomena. In this situation, the use of heuristic models based on the experimental data and experience of technicians is fully justified since, in an approximate manner at least, they allow predicting the mechanical properties of the metal manufactured under given process conditions. The study presents a methodology applicable in the design of a heuristic model based on the formalism of the logic of plausible reasoning (LPR). The problem under consideration consists in finding a technological variant of the process that will give the desired product parameters while minimizing the cost of production. The conducted tests have shown the effectiveness of the proposed approach.

3

Online Supervised Learning Approach for Machine Scheduling

Sądel B., Śnieżyński B.

Schedae Informaticae

|

2016

|

Vol. 25

165--176

EN

Due to rapid growth of computational power and demand for faster and more optimal solution in today's manufacturing, machine learning has lately caught a lot of attention. Thanks to it's ability to adapt to changing conditions in dynamic environments it is perfect choice for processes where rules cannot be explicitly given. In this paper proposes on-line supervised learning approach for optimal scheduling in manufacturing. Although supervised learning is generally not recommended for dynamic problems we try to defeat this conviction and prove it's viable option for this class of problems. Implemented in multi-agent system algorithm is tested against multi-stage, multi-product ow-shop problem. More specically we start from dening considered problem. Next we move to presentation of proposed solution. Later on we show results from conducted experiments and compare our approach to centralized reinforcement learning to measure algorithm performance.

4

A strategy learning model for autonomous agents based on classification

Śnieżyński B.

International Journal of Applied Mathematics and Computer Science

|

2015

|

Vol. 25, no. 3

471--482

EN

In this paper we propose a strategy learning model for autonomous agents based on classification. In the literature, the most commonly used learning method in agent-based systems is reinforcement learning. In our opinion, classification can be considered a good alternative. This type of supervised learning can be used to generate a classifier that allows the agent to choose an appropriate action for execution. Experimental results show that this model can be successfully applied for strategy generation even if rewards are delayed. We compare the efficiency of the proposed model and reinforcement learning using the farmer–pest domain and configurations of various complexity. In complex environments, supervised learning can improve the performance of agents much faster that reinforcement learning. If an appropriate knowledge representation is used, the learned knowledge may be analyzed by humans, which allows tracking the learning process.

5

The use of LPR (logic of plausible reasoning) to obtain information on innovative casting technologies

Kluska-Nawarecka S., Śnieżyński B., Parada W., Lustofin M., Wilk-Kołodziejczyk D.

Archives of Civil and Mechanical Engineering

|

2014

|

Vol. 14, no. 1

25--31

EN

The paper presents a methodology for the application of a formalism of the logic of plausible reasoning (LPR) to create knowledge about a specific problem area. In this case, the methodology has been related to the task of obtaining information about the innovative casting technologies. In the search for documents, formulas created in terms of LPR have a much greater expressive power than the commonly used keywords. The discussion was illustrated with the results obtained using a pilot version of the original information tool.

6

Choice of a knowledge representation method for learning classifiers in medical domains

Śnieżyński B.

Journal of Medical Informatics & Technologies

|

2005

|

Vol. 9

195--200

EN

Knowledge based systems in medical domains are common nowadays. Machine learning techniques are broadly used to generate knowledge for such systems. Developers have to choose not only the learning method, but also, what is even more important, the knowledge representation method. The most common criterion for such a choice is prediction accuracy. In the paper we argue that in certain cases knowledge representation, and its simplicity and intelligibility, are more important. In this paper results of experiments performed using several medical data sets and chosen machine learning algorithms are presented. Next, some examples of learned classifiers are shown. Analysis of results conclude the work.

7

Rule based expert system for hospital infection diagnosis

Śnieżyński B., Valenta M., Zygmunt A.

Journal of Medical Informatics & Technologies

|

2000

|

Vol. 1

CS 9--14

EN

Nowadays, hospital infection diagnosis is a serious problem. The paper deals with expert system for this task. Presented system helps to recognise 12 infection types. It is created using rule based expert system shell Consus. At this paper choice of knowledge representation is mentioned. System architecture is presented, domain knowledge and inference process are characterised.