https://ir.unisa.ac.za/handle/10500/23880 2026-05-09T05:07:11Z 2026-05-09T05:07:11Z Viktor, HL Rennhackkamp, MH https://ir.unisa.ac.za/handle/10500/24042 2018-05-24T01:00:44Z 1992-01-01T00:00:00Z

Network partitions in distributed databases Viktor, HL; Rennhackkamp, MH The technology of distributed databases is based on two other technologies which have developed a solid foundation: computer networks and centralized databases. A distributed database designer should therefore address database-related as well as network-related issues. Provision has to be made for the detection of and recovery from failure of the database as well as the network components. One type of failure resulting from distribution is network partitions. During a network failure the sites of the distributed database are subdivided into two or more disjoint groups. These groups continue processing newly submitted transactions without any knowledge of the operations at the sites in the other partition groups. Since data may be replicated, the various copies of a data item may be updated inconsistently at the different partition groups. When handling continued transaction processing during network partitions, a trade-off is made between the level of availability and the consistency provided by the system. Pessimistic partition processing strategies ensure database consistency by reducing availability. The token lists pessimistic partition processing strategy, which utilizes a reliable broadcasting facility, proved to be the best for the system under consideration.

1992-01-01T00:00:00Z van der Poel, E Cloete, I https://ir.unisa.ac.za/handle/10500/24041 2018-05-24T01:01:03Z 1992-01-01T00:00:00Z

Animating neural network training van der Poel, E; Cloete, I In Artificial Neural Network (ANN) simulation it is usually necessary to examine the behaviour of the ANN and detect problems if they occur. However, due to the large volume and high dimensionality of data generated during ANN simulation, interpretation of results is difficult. By using visualisation techniques, such as simple animated faces, problems that occur during the training of a back-propagation neural network can be detected and analyzed. The dynamic behaviour of the network is also better understood when using animation as part of the visualization process.

1992-01-01T00:00:00Z Tredoux, G https://ir.unisa.ac.za/handle/10500/24040 2018-05-24T01:00:46Z 1992-01-01T00:00:00Z

Mechanising execution sequence semantics in HOL Tredoux, G The mechanization in Higher Order Logic of a general- purpose operational semantics for programming languages is described. The mechanization allows the sound derivation of Dijkstra-style axiomatic semantics. A small programming language fragment is presented as an illustration.

1992-01-01T00:00:00Z Theron, PZ Cloete, I https://ir.unisa.ac.za/handle/10500/24039 2018-05-24T01:01:11Z 1992-01-01T00:00:00Z

Automatically linking words and concepts in an Afrikaans dictionary Theron, PZ; Cloete, I The problem of automatic acquisition of lexical-semantic relations for Afrikaans nouns from a dictionary is addressed. The acquisition process is improved over previous approaches by implementing both typographic constraints in patterns, and by automating the acquisition of syntactic patterns. Results for the noun taxonomy relation IS-A and semantic feature IS-HUMAN show that ambiguity can successfully be resolved with the patterns obtained.

1992-01-01T00:00:00Z