Electronic Theses and Dissertations (ETD)

Permanent URI for this community

https://wiredspace.wits.ac.za/handle/10539/45

Please read important information below and then click on the ETD Collection link

The General Rules of the University requires every post-graduate student to submit two final, corrected copies of the dissertation or research report or thesis in a printed format as well as a final, corrected copy in electronic format.

The availability of Wits theses and dissertations electronically will undoubtedly lead to increased visibility of the institution internationally.

The advantages and attraction of electronic theses and dissertations (ETDs) include the following:

By creating ETDs students can use more innovative approaches, as well as print on paper, including audio, video, hypertext, multimedia and hypermedia.

Access to graduate research information is enhanced by making it available to researchers on the Internet (ETDs are accessed ten to a hundred times more than the paper copy).

Collaboration between research programmes can be promoted at separate universities by making research work visible and accessible via a network archive.

The costs of submitting and handling theses and dissertations will be reduced.

The theses and dissertations will be available 24 hours a day and are never checked out of the library.

For any help with ETD questions you might have, please contact Library Education & Training


ETD Submission Form

Browse

Browsing Electronic Theses and Dissertations (ETD) by Author "Abdella, Mussa Ismael"

Now showing 1 - 1 of 1
  • No Thumbnail Available
    Item
    The use of genetic algorithms and neural networks to approximate missing data in database
    (2006-01-16T07:52:17Z) Abdella, Mussa Ismael
    Missing data creates various problems in analysing and processing of data in databases. Due to this reason missing data has been an area of research in various disciplines for a quite long time. This report intro- duces a new method aimed at approximating missing data in a database using a combination of genetic algorithms and neural networks. The proposed method uses genetic algorithm to minimise an error function derived from an auto-associative neural network. The error function is expressed as the square of the di®erence between the actual observa- tions and predicted values from an auto-associative neural network. In the event of missing data, all the values of the actual observations are not known hence, the error function is decomposed to depend on the known and unknown (missing) values. Multi Layer Perceptron (MLP), and Radial Basis Function (RBF) neural networks are employed to train the neural networks. The research focus also lies on the investigation of using the proposed method in approximating missing data with great accuracy as the number of missing cases within a single record increases. The research also investigates the impact of using di®erent neural net- work architecture in training the neural network and the approximation ii found to the missing values. It is observed that approximations of miss- ing data obtained using the proposed model to be highly accurate with 95% correlation coe±cient between the actual missing values and cor- responding approximated values using the proposed model. It is found that results obtained using RBF are better than MLP. Results found us- ing the combination of both MLP and RBF are found to be better than those obtained using either MLP or RBF. It is also observed that there is no signi¯cant reduction in accuracy of results as the number of missing cases in a single record increases. Approximations found for missing data are also found to depend on the particular neural network architecture employed in training the data set.
Copyright Ownership Is Guided By The University's

Intellectual Property policy

Students submitting a Thesis or Dissertation must be aware of current copyright issues. Both for the protection of your original work as well as the protection of another's copyrighted work, you should follow all current copyright law.