Comparative Analysis Of Knowledge Base System Development Approaches In The Context Of Ethio Telecom

For a successful business, telecom companies need to provide consistent, qualityrnand timely maintenance and support to their customers. This is difficult tornachieve without timely availability of domain experts that can efficiently be managedrnusing the Knowledge Base System (KBS). Different approaches are appliedrnto implement KBS and researchers study accuracy performance of the approaches.rnHowever, there are limited performance studies on the approaches in the contextrnof telecom companies and none for ethio telecom. Most of the studies are alsornonly focused to accuracy performance of the approaches and other relevant performancernmetrics that have an impact on the incident handling time of InformationrnTechnology (IT) experts are not addressed well.rnThis thesis work presents performance comparison for KBS development approachesrnin the context of operation support and maintenance process of ethio telecom.rnRule-based, Case-based and Artificial Neural Network (ANN)-based KBS developmentrnapproaches are considered and these approaches are modeled using datarncollected from ethio telecom technical support experts’ knowledge and experiencernthrough interviews and data analysis. For the performance analysis, data preprocessingrntechniques including data reduction, data cleaning, data integration andrndata transformation are applied on the collected datasets. Execution time, accuracy,rnF-measure, precision, recall, error rate and mean square error metrics thatrnhave an impact on incident handling time are used for the performance evaluation.rnPython programming language is used to model, train, evaluate and comparernthe selected approaches.rnAchieved performance results show that ANN-based KBS development approachrnhas a better performance in terms of building and executing the model, taking arnminimum time of 0.95 seconds. Furthermore, rule-based approach accomplishesrnthe highest accuracy that is 98.6 %, with minimum error in identifying the classrnlabels which are incident types. Based on the evaluation results on the execution time or response time, accuracy, and error measures which have an impact on thernincident handling time, a hybrid KBS development approach from the ANN andrnrule-based KBS development approach can provide the best aggregate result.

Get Full Work

Report copyright infringement or plagiarism