Show simple item record

dc.contributor.authorKagumya, Mark
dc.date.accessioned2018-10-08T19:07:04Z
dc.date.available2018-10-08T19:07:04Z
dc.date.issued2017-01
dc.identifier.urihttp://hdl.handle.net/10570/6562
dc.descriptionA dissertation submitted to the School of Graduate Studies in partial fulfillment of the requirements for the Award of Master of Science in Computer Science Degree of Makerere Universityen_US
dc.description.abstractModeling methods for computer systems currently have functional and feature limitations for secure design. This is because modeling computer systems accurately is complex and requires precision and a variety of features depending on the nature of systems to be designed. Additionally, the increase in technology solutions, frameworks, security threats and other functionality limitations affect performance of modeling methods or techniques. Modeling methods used in computing require improvements in security, user friendliness and dynamism in order to address the limitations. The objective of this research was to design a secure, dynamic and hierarchical modeling method. A literature review is carried out in this research to discuss literature relevant to modeling. The methodology used includes a discussion and evaluation of existing modeling methods and techniques, design of a new optimized modeling method solution and an example case study of a client server network communication model as part of the implementation to show how the new modeling method works. Experiments, validation and results are also presented to verify the implementation as part of the methodology. The solution presented in this research improves computer system modeling as compared to current design methods by creating a secure method for design that uses improved structure, objects, probability and learning. This allows designers to create more organized, hierarchical and descriptive models providing a comprehensive solution for modeling by improving on simplicity, user friendliness and dynamism. As a result various vulnerabilities and attacks were addressed as a result of proper design and probability calculations that allowed learning about security vulnerabilities.en_US
dc.language.isoenen_US
dc.publisherMakerere Universityen_US
dc.subjectModeling computer systemsen_US
dc.subjectArtificial Intelligenceen_US
dc.subjectAuthentication,en_US
dc.subjectBayesian Networks,en_US
dc.titleA modeling method for computer system security using structure and probabilityen_US
dc.typeThesis/Dissertation (Masters)en_US


Files in this item

Thumbnail
Thumbnail

This item appears in the following Collection(s)

Show simple item record