'n Ekspertstelsel vir die beheer van pneumonie in 'n kritiese sorg eenheid.
- Authors: Schoeman, Isabella Lodewina
- Date: 2012-08-14
- Subjects: Expert systems (Computer science) , Medicine - Data processing
- Type: Thesis
- Identifier: uj:9226 , http://hdl.handle.net/10210/5675
- Description: M.Sc. , Surgical patients admitted to an intensive care unit, are susceptible to infection by a large number of micro-organisms. Host defence mechanisms are breached by severe injuries or operations, or the use of life-support systems such as ventilators, catheters and endotracheal tubes. These organisms, some of which are resistant to antibiotics, can therefore invade sterile tissue. Although tissue samples from infected sites are sent to a laboratory to be analyzed, treatment of the patient has to commence before the results are known. Intelligent computer systems, of which expert systems are one of the most popular applications, can be utilized to support diagnostic and therapeutical decisions. This thesis describes the development of an expert system that supports clinical decision-making in the diagnosis and treatment of hospital-aquired pneumonia in an intensive care unit. Input data required by the expert system module are extracted from a data base with patient records. The data base and expert system module communicates by means of a program written in a conventional programming language. The system, which is only a prototype, can be extended to include additional expert system modules addressing other infections. Aquiring knowledge to be encoded in the expert system's knowledge base, remains a problem. In this case an existing scoring system that assigns weights to measurements and the outcomes of certain investigations, is used to obtain a score according to which pneumonia can be diagnosed. The infection is subsequently classified as one of several categories, according to existing guidelines. Appropriate therapy is recommended. The system can also consult a file containing sensitivities of bacteria for antibiotics for the unit, in order to facilitate the choice of drugs. The system has been implemented and tested with a few cases.
- Full Text:
- Authors: Schoeman, Isabella Lodewina
- Date: 2012-08-14
- Subjects: Expert systems (Computer science) , Medicine - Data processing
- Type: Thesis
- Identifier: uj:9226 , http://hdl.handle.net/10210/5675
- Description: M.Sc. , Surgical patients admitted to an intensive care unit, are susceptible to infection by a large number of micro-organisms. Host defence mechanisms are breached by severe injuries or operations, or the use of life-support systems such as ventilators, catheters and endotracheal tubes. These organisms, some of which are resistant to antibiotics, can therefore invade sterile tissue. Although tissue samples from infected sites are sent to a laboratory to be analyzed, treatment of the patient has to commence before the results are known. Intelligent computer systems, of which expert systems are one of the most popular applications, can be utilized to support diagnostic and therapeutical decisions. This thesis describes the development of an expert system that supports clinical decision-making in the diagnosis and treatment of hospital-aquired pneumonia in an intensive care unit. Input data required by the expert system module are extracted from a data base with patient records. The data base and expert system module communicates by means of a program written in a conventional programming language. The system, which is only a prototype, can be extended to include additional expert system modules addressing other infections. Aquiring knowledge to be encoded in the expert system's knowledge base, remains a problem. In this case an existing scoring system that assigns weights to measurements and the outcomes of certain investigations, is used to obtain a score according to which pneumonia can be diagnosed. The infection is subsequently classified as one of several categories, according to existing guidelines. Appropriate therapy is recommended. The system can also consult a file containing sensitivities of bacteria for antibiotics for the unit, in order to facilitate the choice of drugs. The system has been implemented and tested with a few cases.
- Full Text:
'n Vergelykende studie van kennisvoorstelling in ekspertstelsels met spesifieke verwysing na die toepassing van die formeletaal-teorie
- Authors: Enslin, Daniel Jacobus
- Date: 2014-09-15
- Subjects: Expert systems (Computer science)
- Type: Thesis
- Identifier: uj:12290 , http://hdl.handle.net/10210/12076
- Description: M.Com. (Informatics) , Please refer to full text to view abstract
- Full Text:
- Authors: Enslin, Daniel Jacobus
- Date: 2014-09-15
- Subjects: Expert systems (Computer science)
- Type: Thesis
- Identifier: uj:12290 , http://hdl.handle.net/10210/12076
- Description: M.Com. (Informatics) , Please refer to full text to view abstract
- Full Text:
A methodology for the implementation of expert systems in large organizations
- Authors: Klut, Johannes Pieter
- Date: 2014-08-19
- Subjects: Expert systems (Computer science) , Business information services - Data processing , Business - Databases
- Type: Thesis
- Identifier: uj:12101 , http://hdl.handle.net/10210/11846
- Description: M.Com. (Informatics) , Please refer to full text to view abstract
- Full Text:
- Authors: Klut, Johannes Pieter
- Date: 2014-08-19
- Subjects: Expert systems (Computer science) , Business information services - Data processing , Business - Databases
- Type: Thesis
- Identifier: uj:12101 , http://hdl.handle.net/10210/11846
- Description: M.Com. (Informatics) , Please refer to full text to view abstract
- Full Text:
A neural-network/expert system based approach for design improvements of products manufactured by the EDM process
- Authors: Naude, Johannes Jacobus
- Date: 2014-09-11
- Subjects: Expert systems (Computer science) , Neural networks (Computer science) , Electro Discharge Machining
- Type: Thesis
- Identifier: uj:12273 , http://hdl.handle.net/10210/12036
- Description: M.Ing. (Mechanical Engineering) , Please refer to full text to view abstract
- Full Text:
- Authors: Naude, Johannes Jacobus
- Date: 2014-09-11
- Subjects: Expert systems (Computer science) , Neural networks (Computer science) , Electro Discharge Machining
- Type: Thesis
- Identifier: uj:12273 , http://hdl.handle.net/10210/12036
- Description: M.Ing. (Mechanical Engineering) , Please refer to full text to view abstract
- Full Text:
An implementation framework for knowledge-based engineering projects
- Authors: Mvudi, Yannick
- Date: 2013-05-27
- Subjects: Knowledge-based engineering , Product design - Data processing , CAD/CAM systems , Expert systems (Computer science)
- Type: Thesis
- Identifier: http://ujcontent.uj.ac.za8080/10210/373562 , uj:7543 , http://hdl.handle.net/10210/8401
- Description: M.Ing. (Engineering Management) , The growing need for customized solutions and faster product delivery obliges the product development industry to develop new strategies that can enable the rapid and flexible design of products. Several design approaches have been developed to address this issue: one such is Knowledge-Based Engineering (KBE), which is a design technique that enables the automation of the design process. This approach consists of using computational intelligence to capture the design rules related to a product family in order to generate automatically customized designs adapted to particular customer requirements. Knowledge-Based Engineering is also used to facilitate the performance of design evaluation activities such as finite element analysis (FEA) and computational fluid dynamics (CFD) as part of multi-disciplinary design optimization (MDO). The application of this approach led to impressive results mostly in the automotive and aeronautical industry. Owing to this method, some companies manage to reduce the duration of the design process by 90%. Despite the excellent results obtained through the use of Knowledge-Based Engineering, there are still very few companies that make use of this approach in their design process. The review of the relevant literature showed that the lack of a standard easy-to-use methodology of implementation is one of the major obstacles to the expansion of Knowledge-Based Engineering. The knowledge processing phase constitutes one of the main challenges of the KBE implementation process. This phase consists of extracting and documenting the knowledge embedded in the design team in order to convert it in a programming code. Available methodologies such as MOKA and KNOMAD do not seem to provide easy-to-use methods to represent the design knowledge in a form that makes it easy to be programmed. The lack of a preliminary stage that justifies the adequacy of KBE for a particular design process is also an important gap identified in the literature.This dissertation discusses a detailed method that addresses issues related to knowledge processing and suitability analysis in KBE implementation. The knowledge processing method suggested is based on the Work Breakdown Structure (WBS) which is used widely in the system engineering approach and consists of a very logical classification of the design knowledge. The strength of this method lies in its ability to represent the design knowledge in a form that makes it understandable for both engineers and programmers. Appropriate representation of this sort shortens the duration of the knowledge processing and facilitates the knowledge programming phase. Regarding the rationale for choosing of KBE, a detailed suitability assessment method is proposed.
- Full Text:
- Authors: Mvudi, Yannick
- Date: 2013-05-27
- Subjects: Knowledge-based engineering , Product design - Data processing , CAD/CAM systems , Expert systems (Computer science)
- Type: Thesis
- Identifier: http://ujcontent.uj.ac.za8080/10210/373562 , uj:7543 , http://hdl.handle.net/10210/8401
- Description: M.Ing. (Engineering Management) , The growing need for customized solutions and faster product delivery obliges the product development industry to develop new strategies that can enable the rapid and flexible design of products. Several design approaches have been developed to address this issue: one such is Knowledge-Based Engineering (KBE), which is a design technique that enables the automation of the design process. This approach consists of using computational intelligence to capture the design rules related to a product family in order to generate automatically customized designs adapted to particular customer requirements. Knowledge-Based Engineering is also used to facilitate the performance of design evaluation activities such as finite element analysis (FEA) and computational fluid dynamics (CFD) as part of multi-disciplinary design optimization (MDO). The application of this approach led to impressive results mostly in the automotive and aeronautical industry. Owing to this method, some companies manage to reduce the duration of the design process by 90%. Despite the excellent results obtained through the use of Knowledge-Based Engineering, there are still very few companies that make use of this approach in their design process. The review of the relevant literature showed that the lack of a standard easy-to-use methodology of implementation is one of the major obstacles to the expansion of Knowledge-Based Engineering. The knowledge processing phase constitutes one of the main challenges of the KBE implementation process. This phase consists of extracting and documenting the knowledge embedded in the design team in order to convert it in a programming code. Available methodologies such as MOKA and KNOMAD do not seem to provide easy-to-use methods to represent the design knowledge in a form that makes it easy to be programmed. The lack of a preliminary stage that justifies the adequacy of KBE for a particular design process is also an important gap identified in the literature.This dissertation discusses a detailed method that addresses issues related to knowledge processing and suitability analysis in KBE implementation. The knowledge processing method suggested is based on the Work Breakdown Structure (WBS) which is used widely in the system engineering approach and consists of a very logical classification of the design knowledge. The strength of this method lies in its ability to represent the design knowledge in a form that makes it understandable for both engineers and programmers. Appropriate representation of this sort shortens the duration of the knowledge processing and facilitates the knowledge programming phase. Regarding the rationale for choosing of KBE, a detailed suitability assessment method is proposed.
- Full Text:
Audit considerations in neural net based knowledge systems
- Authors: Oosthuizen, Wouter
- Date: 2014-05-08
- Subjects: Expert systems (Computer science) , Neural networks (Computer science) , Audutung - Data processing
- Type: Thesis
- Identifier: uj:10956 , http://hdl.handle.net/10210/10529
- Description: M.Com. (Computer Auditing) , Knowledge-based systems (KBS's) is one of the new and emerging technologies, and it is expected to impact dramatically on the information systems of the 1990's. As many of these systems impact business operations and have significant financial influence, auditors should be involved in the implementation of this emerging technology and should review it for the presence and continued operation of effective controls. Because this technology up to now has been tried only in limited environments, the relevant risk, control and audit issues are not always clearly defined. The objective of this essay will be to identify the risks, controls and audit considerations in a knowledge-based system environment, with special reference to neural networks. As neural networks have got the ability to modify themselves without any human intervention, this can have major audit implications - a topic which has not been dealt with before by the audit community.
- Full Text:
- Authors: Oosthuizen, Wouter
- Date: 2014-05-08
- Subjects: Expert systems (Computer science) , Neural networks (Computer science) , Audutung - Data processing
- Type: Thesis
- Identifier: uj:10956 , http://hdl.handle.net/10210/10529
- Description: M.Com. (Computer Auditing) , Knowledge-based systems (KBS's) is one of the new and emerging technologies, and it is expected to impact dramatically on the information systems of the 1990's. As many of these systems impact business operations and have significant financial influence, auditors should be involved in the implementation of this emerging technology and should review it for the presence and continued operation of effective controls. Because this technology up to now has been tried only in limited environments, the relevant risk, control and audit issues are not always clearly defined. The objective of this essay will be to identify the risks, controls and audit considerations in a knowledge-based system environment, with special reference to neural networks. As neural networks have got the ability to modify themselves without any human intervention, this can have major audit implications - a topic which has not been dealt with before by the audit community.
- Full Text:
Design and implementation of a supervisory expert system for hot rolling process optimisation
- Authors: Jordaan, Albertus Laing
- Date: 2009-02-26T12:20:29Z
- Subjects: Expert systems (Computer science) , Process control , Stainless steel industry
- Type: Mini-Dissertation
- Identifier: http://ujcontent.uj.ac.za8080/10210/367454 , uj:8162 , http://hdl.handle.net/10210/2168
- Description: M.Ing. , This dissertation stresses the need and value of adding an Artificial Intelligence level between manufacturing units and First Principle Models that supply processing set points. The value is proven by evaluating the effect of designing and implementing a supervisory expert system for hot rolling process optimisation at Columbus Stainless Ltd. Pty. The hot rolling mill’s mainly fuzzy logic artificial intelligence “Expert System” functions as an extra diagnostic and control system that manages the performance of the processing set point models. The Expert System was developed in order to effectively imitate what the human experts used to do – which was to virtually continuously optimise the rolling process by making database data changes. The human experts had to make these process adjustments in order to compensate for the unavoidable imperfections and shortcomings of First Principle Models, which are unable to perfectly model nature. Even though the Expert System addresses a vast amount of hot rolling product quality and throughput control aspects, it still sufficiently automates human expert control to successfully manage production performance in all areas of control. This document was written in a format that would comply with Masters Degree Dissertation standards, while providing a document that can easily be used by Columbus Stainless personnel as a reference book of the philosophies, strategies and design of the Expert System.
- Full Text:
- Authors: Jordaan, Albertus Laing
- Date: 2009-02-26T12:20:29Z
- Subjects: Expert systems (Computer science) , Process control , Stainless steel industry
- Type: Mini-Dissertation
- Identifier: http://ujcontent.uj.ac.za8080/10210/367454 , uj:8162 , http://hdl.handle.net/10210/2168
- Description: M.Ing. , This dissertation stresses the need and value of adding an Artificial Intelligence level between manufacturing units and First Principle Models that supply processing set points. The value is proven by evaluating the effect of designing and implementing a supervisory expert system for hot rolling process optimisation at Columbus Stainless Ltd. Pty. The hot rolling mill’s mainly fuzzy logic artificial intelligence “Expert System” functions as an extra diagnostic and control system that manages the performance of the processing set point models. The Expert System was developed in order to effectively imitate what the human experts used to do – which was to virtually continuously optimise the rolling process by making database data changes. The human experts had to make these process adjustments in order to compensate for the unavoidable imperfections and shortcomings of First Principle Models, which are unable to perfectly model nature. Even though the Expert System addresses a vast amount of hot rolling product quality and throughput control aspects, it still sufficiently automates human expert control to successfully manage production performance in all areas of control. This document was written in a format that would comply with Masters Degree Dissertation standards, while providing a document that can easily be used by Columbus Stainless personnel as a reference book of the philosophies, strategies and design of the Expert System.
- Full Text:
Die toepassing van ekspertstelseltegnologie binne inligtingsekerheid
- Authors: De Ru, Willem Gerhardus
- Date: 2014-09-18
- Subjects: Expert systems (Computer science) , Data protection , Data security
- Type: Thesis
- Identifier: http://ujcontent.uj.ac.za8080/10210/372918 , uj:12339 , http://hdl.handle.net/10210/12125
- Description: M.Sc. (Computer Science) , Please refer to full text to view abstract
- Full Text:
- Authors: De Ru, Willem Gerhardus
- Date: 2014-09-18
- Subjects: Expert systems (Computer science) , Data protection , Data security
- Type: Thesis
- Identifier: http://ujcontent.uj.ac.za8080/10210/372918 , uj:12339 , http://hdl.handle.net/10210/12125
- Description: M.Sc. (Computer Science) , Please refer to full text to view abstract
- Full Text:
Evaluering van strategiese markinligting met behulp van 'n ekspertbemarkingsbeplanningstelsel
- Authors: Burger, André
- Date: 2014-04-14
- Subjects: Strategic planning - Data processing , Marketing - Management , Expert systems (Computer science)
- Type: Thesis
- Identifier: uj:10598 , http://hdl.handle.net/10210/10121
- Description: M.Comm. (Business Management) , The evaluation of strategic market information with the aid of an expert marketing planning system. Market information is one of the key inputs to a strategic marketing plan. The display of the market information with the aid of an expert marketing planning system will be discussed in this thesis. Market information was obtained by means of a questionnaire. The information was analysed and the marketing planning expert system was used to evaluate the marketing opportunities. The evaluation of the expert marketing planning system was done against ten points for a marketing planning process. The conclusion of this thesis is that an expert marketing planning system can aid the marketing planner in the evaluation of the strategic market information. However it is important to manage the planning process to obtain a marketing plan that will be accepted and implemented by the organisation.
- Full Text:
- Authors: Burger, André
- Date: 2014-04-14
- Subjects: Strategic planning - Data processing , Marketing - Management , Expert systems (Computer science)
- Type: Thesis
- Identifier: uj:10598 , http://hdl.handle.net/10210/10121
- Description: M.Comm. (Business Management) , The evaluation of strategic market information with the aid of an expert marketing planning system. Market information is one of the key inputs to a strategic marketing plan. The display of the market information with the aid of an expert marketing planning system will be discussed in this thesis. Market information was obtained by means of a questionnaire. The information was analysed and the marketing planning expert system was used to evaluate the marketing opportunities. The evaluation of the expert marketing planning system was done against ten points for a marketing planning process. The conclusion of this thesis is that an expert marketing planning system can aid the marketing planner in the evaluation of the strategic market information. However it is important to manage the planning process to obtain a marketing plan that will be accepted and implemented by the organisation.
- Full Text:
Ontwikkeling van koste-effektiewe ekspertstelsels
- Authors: Du Plessis, Gerhard Michael
- Date: 2012-08-20
- Subjects: Expert systems (Computer science) - Cost effectiveness , Expert systems (Computer science)
- Type: Thesis
- Identifier: uj:2747 , http://hdl.handle.net/10210/6187
- Description: M.Comm. , The present study comprises an investigation into and a discussion on the development of expert systems. The theme of the study is of topical interest, since expert systems form an important part of the latest developments in the domain of computer technology. At present, a general deficiency is being experienced in the computer industry in respect of a structured development methodology, and specifically in respect of cost-effective expert systems. The reasons for and causes of the said deficiency, as well as the different ways and means in which possibly to supply this deficiency, represent the issue to be investigated in the present study. Owing to the above-mentioned deficiency in respect of information system methodologies, certain applications and principles have been used throughout the study in a bid to illustrate the differences and similarities between the developmental aspects of certain conventional systems, compared to the same developmental aspects of expert systems. The importance of the various stages of development that the expert system undergo also comes under consideration, together with the development applications and resources, the criteria for cost effectiveness, knowledge engineering and information security. Hereafter, the Sphere Model, based on the introductory chapters, is presented as the solution to the problem posed in the study. This model can, for the entire life cycle of an expert system, be applied in respect of the analysis of the cost effectiveness of such system, and can, at the same time, assist in gathering vital information on the production of subsequent highquality and cost-effective expert systems.
- Full Text:
- Authors: Du Plessis, Gerhard Michael
- Date: 2012-08-20
- Subjects: Expert systems (Computer science) - Cost effectiveness , Expert systems (Computer science)
- Type: Thesis
- Identifier: uj:2747 , http://hdl.handle.net/10210/6187
- Description: M.Comm. , The present study comprises an investigation into and a discussion on the development of expert systems. The theme of the study is of topical interest, since expert systems form an important part of the latest developments in the domain of computer technology. At present, a general deficiency is being experienced in the computer industry in respect of a structured development methodology, and specifically in respect of cost-effective expert systems. The reasons for and causes of the said deficiency, as well as the different ways and means in which possibly to supply this deficiency, represent the issue to be investigated in the present study. Owing to the above-mentioned deficiency in respect of information system methodologies, certain applications and principles have been used throughout the study in a bid to illustrate the differences and similarities between the developmental aspects of certain conventional systems, compared to the same developmental aspects of expert systems. The importance of the various stages of development that the expert system undergo also comes under consideration, together with the development applications and resources, the criteria for cost effectiveness, knowledge engineering and information security. Hereafter, the Sphere Model, based on the introductory chapters, is presented as the solution to the problem posed in the study. This model can, for the entire life cycle of an expert system, be applied in respect of the analysis of the cost effectiveness of such system, and can, at the same time, assist in gathering vital information on the production of subsequent highquality and cost-effective expert systems.
- Full Text:
The normalization of frames as a superclass of relations
- Authors: Jonker, Jacob
- Date: 2014-11-17
- Subjects: Relational databases , Expert systems (Computer science)
- Type: Thesis
- Identifier: uj:12839 , http://hdl.handle.net/10210/12728
- Description: M.Sc. (Computer science) , Knowledge representation suffers from certain problems, which is not a result of the inadequacies of knowledge representation schemes, but of the way in which they are used and implemented. In the first part of this dissertation we examine the relational model (as used in relational database management systems) and we examine frames (a knowledge representation scheme used in expert systems), as proposed by M. Minsky [MIN75]. We then provide our own definition of frames. In the second part, we examine similarities between the two models (the relational model and our frame model), establishing frames as a superclass of relations. We then define normalization for frames and examine how normalization might solve some of the problems we have identified. We then examine the integration of knowledge-based systems and database management systems and classify our normalization of frames as such an attempt. We conclude by examining the place of normalization within the expert system development life cycle
- Full Text:
- Authors: Jonker, Jacob
- Date: 2014-11-17
- Subjects: Relational databases , Expert systems (Computer science)
- Type: Thesis
- Identifier: uj:12839 , http://hdl.handle.net/10210/12728
- Description: M.Sc. (Computer science) , Knowledge representation suffers from certain problems, which is not a result of the inadequacies of knowledge representation schemes, but of the way in which they are used and implemented. In the first part of this dissertation we examine the relational model (as used in relational database management systems) and we examine frames (a knowledge representation scheme used in expert systems), as proposed by M. Minsky [MIN75]. We then provide our own definition of frames. In the second part, we examine similarities between the two models (the relational model and our frame model), establishing frames as a superclass of relations. We then define normalization for frames and examine how normalization might solve some of the problems we have identified. We then examine the integration of knowledge-based systems and database management systems and classify our normalization of frames as such an attempt. We conclude by examining the place of normalization within the expert system development life cycle
- Full Text:
- «
- ‹
- 1
- ›
- »