Thesis (Ph.D.)--University of Adelaide, Dept. of Environmental Studies, 2003. / Bibliography: leaves 281-330.

Thesis (Ph.D.)--University of Delaware, 2005. / Principal faculty advisor: Babatunde Ogunnaike, Dept. of Chemical Engineering. Includes bibliographical references.

Thesis (M.S.)--Brandeis University, 2009. / Title from PDF title page (viewed on May 29, 2009). Includes bibliographical references.

Thesis (Master of Health Policy and Administration)--Washington State University, May 2009. / Title from PDF title page (viewed on June 5, 2009). "Department of Health Policy and Administration." Includes bibliographical references (p. 34-39).

No description available.

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.

Summary in Chinese. / Thesis (M.S.)--Chinese University of Hong Kong. / Bibliography: l. 84-95.

No description available.

This work examines the impact that genes with activation relationships have on Genetic Algorithms (GAs). These activation relationships allow genes to control whether other associated genes get expressed in the phenotype. More specifically, this thesis investigates the effect that the incorporation of control levels (tiers of genes which determine the activation of lower genes) have on GAs which are used to generate feed forward Artificial Neural Networks (ANNs). In order to evaluate the performance of different levelled Structured Genetic Algorithms (SGAs; they are GAs which possess control levels), numerous experiments were conducted, utilising the four input XOR, Mackey-Glass and Breast Cancer data sets. In addition, the thesis derives four mathematical models which describe how SGA redundancy changes as more control levels are incorporated. This thesis also presented and implemented a novel four level Structurally Evolved Neural Network Algorithm (SENNGA). Furthermore, it demonstrated the novel implementation of a three level SENNGA. Empirical results show that the incorporation of control levels can improve convergence speed and accuracy, up until the optimum number of control levels is reached. In all experiments increasing the number of control levels had the effect of encouraging the production of leaner and more efficient ANNs. Three and four level SENNGAs can demonstrate superior performance in terms of reducing the training error and generalisation. However this does not apply in all cases. Of particular note, three and four level SENNGAs have exhibited a reduced level of generalisation in the breast cancer experiments, compared two level ones. / Doctor of Philosophy (PhD)

PURPOSE: The Center for Minority Health (CMH) in the University of Pittsburghs Graduate School of Public Health established The Healthy Black Family Project, a program designed to increase awareness of the contribution of family health history to the development of chronic diseases. We assessed the impact of a family health history session on African Americans risk perceptions for the development of chronic diseases, which result from interactions between genes and the environment. The public health significance of this study was to delineate how participants perceived risks for developing chronic diseases (i.e., cancer, cardiovascular disease, etc.) would shape risk-reducing behavior modifications and utilization of preventive services. METHODS: Participants (n=175) completed interviews to create a family health history (or pedigree), a schematic representation of health history information in a family. Of these individuals, a total of 125 participants completed surveys that assessed their perceptions of risk for nine chronic diseases. For the purpose of this study, statistical analysis was limited to colorectal cancer (CRC) and cardiovascular disease (CVD). Assessments of risk perception before and following the family health history sessions were calculated to assess changes in accuracy of risk. RESULTS: Overall, participants appeared to understand the contribution of general risk factors (i.e., smoking) to disease development. However, participants were less knowledgeable about risk related to family health history. Of the 125 participants, sixty-nine percent (n=86) and eighty-five percent (n=107) overestimated the lifetime risks to develop colon cancer for women and men in the general population, respectively. Similar trends were observed for heart disease. More participants were accurate about their risk perceptions for colon cancer than for heart disease in both the pre- and post-family health history session. Among the participants whose perceptions changed, inaccurate perceptions for colon cancer and heart disease prior to the family health history interview were significantly more likely to become accurate for colon cancer (p=0.028) and heart disease (p=0.005). CONCLUSIONS: The family health history is an effective tool in identifying at-risk individuals and promoting accurate risk perceptions. Encouraging the use of family health history and providing accurate risk perceptions can lead to healthy behavior modifications that may decrease racial and ethnic health disparities.