Developing an Expert System for Computer and Network Troubleshooting

Amanuel, Ayde Ergado (2014) Developing an Expert System for Computer and Network Troubleshooting. Masters thesis, Addis Ababa University.

PDF (Developing an Expert System for Computer and Network Troubleshooting) Amanuel, Ayde Ergado.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Information and communication technology become crucial in any aspects of organizations activity to perform their tasks. However, computer and network troubleshooting services were not efficient in developing countries like Ethiopia as it is required to be. In Ethiopia, computer related problems are hot issues in different organizations those were using computer and computer related devices to perform their tasks. In computer domain the professionals were limited in number but the number of institutions looking for computer professionals was high. The major challenges in troubleshooting computer and network are lack of expertise to troubleshoot problems with hardware and software which is leading to problems such as connection and virus attaches. These and other computer related problems need maintenance from the ICT technicians. But the problems are not solved as required by the computer users due to different reasons (skill level of ICT technicians, limited number of ICT technical experts, turnover rate and motivation of ICT professionals). As a result of this, the main aim of this study is developing knowledge based expert system which is providing troubleshooting information about problems occurred in the computer system for the ICT technicians and computer users to solve problems effectively and efficiently to utilize computer and computer related resources. To achieve this objective, domain knowledge was acquired using semi-structured interview technique, observation and document analysis. Domain experts were selected from Jimma University ICT office using purposive sampling techniques. The conceptual model of the knowledge based system is designed by using a decision tree structure which is easy to understand and interpret the causes involved in computer troubleshooting. Based on the conceptual model, the prototype was developed with SWI WIN prolog by using ‗if – then‘ rules. The developed prototype uses backward chaining to infer the rules and provide appropriate recommendations. The prototype was evaluated by using different techniques and the performance of the prototype knowledge based system had got good acceptance by the system evaluators. According to the system evaluators 83.6% of the users were satisfied with the prototype. In addition, the performance of the system was evaluated by using predictive validation techniques with twelve test cases. The results of the validation test cases indicate that the prototype registers on the average 75% accuracy. The system is learning its facts but not fully update rules as well, which left for further research.

Item Type:	Thesis (Masters)
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software
Divisions:	Africana
Depositing User:	Selom Ghislain
Date Deposited:	19 Jun 2018 14:28
Last Modified:	19 Jun 2018 14:28
URI:	http://thesisbank.jhia.ac.ke/id/eprint/4693

Actions (login required)

View Item