Show simple item record

dc.contributor.authorAtampore, Francis
dc.date.accessioned2012-09-18T13:42:32Z
dc.date.available2012-09-18T13:42:32Z
dc.date.issued2012-09-18
dc.identifier.urihttp://hdl.handle.net/10464/4102
dc.description.abstractRelation algebras is one of the state-of-the-art means used by mathematicians and computer scientists for solving very complex problems. As a result, a computer algebra system for relation algebras called RelView has been developed at Kiel University. RelView works within the standard model of relation algebras. On the other hand, relation algebras do have other models which may have different properties. For example, in the standard model we always have L;L=L (the composition of two (heterogeneous) universal relations yields a universal relation). This is not true in some non-standard models. Therefore, any example in RelView will always satisfy this property even though it is not true in general. On the other hand, it has been shown that every relation algebra with relational sums and subobjects can be seen as matrix algebra similar to the correspondence of binary relations between sets and Boolean matrices. The aim of my research is to develop a new system that works with both standard and non-standard models for arbitrary relations using multiple-valued decision diagrams (MDDs). This system will implement relations as matrix algebras. The proposed structure is a library written in C which can be imported by other languages such as Java or Haskell.en_US
dc.language.isoengen_US
dc.publisherBrock Universityen_US
dc.subjectRelationsen_US
dc.subjectROMDDen_US
dc.subjectMatrix Algebraen_US
dc.titleRelMDD-A Library for Manipulating Relations Based on MDDsen_US
dc.typeElectronic Thesis or Dissertationen
dc.degree.nameM.Sc. Computer Scienceen_US
dc.degree.levelMastersen_US
dc.contributor.departmentDepartment of Computer Scienceen_US
dc.degree.disciplineFaculty of Mathematics and Scienceen_US
refterms.dateFOA2021-08-08T02:22:36Z


Files in this item

Thumbnail
Name:
Brock_Atampore_Francis_2012.pdf
Size:
1023.Kb
Format:
PDF

This item appears in the following Collection(s)

Show simple item record