Thesis(M.S.)--Auburn University, 2005. / Abstract. Vita. Includes bibliographic references.

Thesis (M.C.S.)--Carleton University, 2004. / Includes bibliographical references (p. 84-91). Also available in electronic format on the Internet.

Thesis (M.S.)--Bowling Green State University, 2009. / Document formatted into pages; contains xi, 81 p. : ill., maps. Includes bibliographical references.

Feature engineering is a process that augments the feature vector of a machine learning model with calculated values that are designed to enhance the accuracy of a model’s predictions. Research has shown that the accuracy of models such as deep neural networks, support vector machines, and tree/forest-based algorithms sometimes benefit from feature engineering. Expressions that combine one or more of the original features usually create these engineered features. The choice of the exact structure of an engineered feature is dependent on the type of machine learning model in use. Previous research demonstrated that various model families benefit from different types of engineered feature. Random forests, gradient-boosting machines, or other tree-based models might not see the same accuracy gain that an engineered feature allowed neural networks, generalized linear models, or other dot-product based models to achieve on the same data set. This dissertation presents a genetic programming-based algorithm that automatically engineers features that increase the accuracy of deep neural networks for some data sets. For a genetic programming algorithm to be effective, it must prioritize the search space and efficiently evaluate what it finds. This dissertation algorithm faced a potential search space composed of all possible mathematical combinations of the original feature vector. Five experiments were designed to guide the search process to efficiently evolve good engineered features. The result of this dissertation is an automated feature engineering (AFE) algorithm that is computationally efficient, even though a neural network is used to evaluate each candidate feature. This approach gave the algorithm a greater opportunity to specifically target deep neural networks in its search for engineered features that improve accuracy. Finally, a sixth experiment empirically demonstrated the degree to which this algorithm improved the accuracy of neural networks on data sets augmented by the algorithm’s engineered features.

Theme of this thesis is practical realization of so-called Island model which is one of way of parallel genetic programming. First part is theoretical. This part is describing terms of genetic programming, age-layered population structure and island model. In second part of thesis is described realization of island model in Java language.

The aim of this diploma thesis is using cartesian genetic programming on design image filters and creating basic structure for implement diferent type of problems. Genetic programming is rapidly growing method, which often using for solve dificult problems. This thesis analyze basic principle, way of application and implementing this method to design filters. Result of this thesis is program realize design filters define by specific parameters, overview of implementig method and achieve summary from this sphere.

Bankruptcy is a highly significant worldwide problem with high social costs. Traditional bankruptcy risk models have been criticized for falling short with respect to bankruptcy theory building due to either modeling assumptions or model complexity. Genetic programming minimizes the amount of a priori structure that is associated with traditional functional forms and statistical selection procedures, but still produces easily understandable and implementable models. Genetic programming was used to analyze 28 potential bankruptcy variables found to be significant in multiple prior research studies, including 10 fraud risk factors. Data was taken from a sample of 422 bankrupt and non-bankrupt Norwegian companies for the period 1993-1998. Six variables were determined to be significant. A genetic programming model was developed for the six variables from an expanded sample of 1136 bankrupt and non-bankrupt Norwegian companies. The model was 81% accurate on a validation sample, slightly better than prior genetic programming research on US public companies, and statistically significantly better than the 77% accuracy of a traditional logit model developed using the same variables and data. The most significant variable in the final model was the prior auditor opinion, thus validating the information value of the auditor's report. The model provides insight into the complex interaction of bankruptcy related factors, especially the effect of company size. The results suggest that accounting information, including the auditor's evaluation of it, is more important for larger than smaller firms. It also suggests that for small firms the most important information is liquidity and non-accounting information. The genetic programming model relationships developed in this study also support prior bankruptcy research, including the finding that company size decreases bankruptcy risk when profits are positive. It also confirms that very high profit levels are associated with increased bankruptcy risk even for large companies an association that may be reflecting the potential for management to be "Cooking the Books".

Unbonded post-tensioned concrete block (UPCB) shear walls are an effective seismic force resisting system due to their ability to contain expected damage attributed to their self-centering capabilities. A few design procedures were proposed to predict the in-plane flexural response of UPCB walls, albeit following only basic mechanics and/or extensive iterative methods. Such procedures, however, may not be capable of capturing the complex nonlinear relationships between different parameters that affect UPCB walls’ behavior or are tedious to be adopted for design practice. In addition, the limited datasets used to validate these procedures may render their accuracy and generalizability questionable, further hindering their adoption by practitioners and design standards. To address these issues, an experimentally-validated nonlinear numerical model was adopted in this study and subsequently employed to simulate 95 UPCB walls with different design parameters to compensate for the lack of relevant experimental data in the current literature. Guided by mechanics and using this database, an evolutionary algorithm, multigene genetic programming (MGGP), was adopted to uncover the relationships controlling the response of UPCB walls, and subsequently develop simplified closed-form wall behavior prediction expressions. Specifically, through integrating MGGP and basic mechanics, a penta-linear backbone model was developed to predict the load-displacement backbone for UPCB walls up to 20% strength degradation. Compared to existing predictive procedures, the prediction accuracy of the developed model and its closed-form nature are expected to enable UPCB wall adoption by seismic design standards and code committees. / Thesis / Master of Applied Science (MASc)

This thesis presents Genetic Programming methodologies to find successful and understandable technical trading rules for financial markets. The methods when applied to the S&P500 consistently beat the buy-and-hold strategy over a 12-year period, even when considering transaction costs. Some of the methods described discover rules that beat the S&P500 with 99% significance. The work describes the use of a complexity-penalizing factor to avoid overfitting and improve comprehensibility of the rules produced by GPs. The effect of this factor on the returns for this domain area is studied and the results indicated that it increased the predictive ability of the rules. A restricted set of operators and domain knowledge were used to improve comprehensibility. In particular, arithmetic operators were eliminated and a number of technical indicators in addition to the widely used moving averages, such as trend lines and local maxima and minima were added. A new evaluation function that tests for consistency of returns in addition to total returns is introduced. Different cooperative coevolutionary genetic programming strategies for improving returns are studied and the results analyzed. We find that paired collaborator coevolution has the best results.

[pt] As técnicas de programação genética permitem flexibilidade no processo de otimização, possibilitando sua aplicação em diferentes áreas do conhecimento e fornecendo novas maneiras para que especialistas avancem em suas áreas com mais rapidez. Parameter mapping approach é um método de otimização numérica que utiliza a programação genética para mapear valores iniciais em parâmetros ótimos para um sistema. Embora esta abordagem produza bons resultados para problemas com soluções triviais, o uso de grandes equações/árvores pode ser necessário para tornar este mapeamento apropriado em sistemas mais complexos.A fim de aumentar a flexibilidade e aplicabilidade do método a sistemas de diferentes níveis de complexidade, este trabalho introduz uma generalização utilizando a programação genética multigênica, para realizar um mapeamento multivariado, evitando grandes estruturas complexas. Foram considerados três conjuntos de funções de benchmark, variando em complexidade e dimensionalidade. Análises estatísticas foram realizadas, sugerindo que este novo método é mais flexível e mais eficiente (em média), considerando funções de benchmark complexas e de grande dimensionalidade. Esta tese também apresenta uma abordagem do novo algoritmo para otimização numérica multimodal.Este segundo algoritmo utiliza algumas técnicas de niching, baseadas no procedimento chamado de clearing, para manter a diversidade da população. Um conjunto benchmark de funções multimodais, com diferentes características e níveis de dificuldade,foi utilizado para avaliar esse novo algoritmo. A análise estatística sugeriu que esse novo método multimodal, que também utiliza programação genética multigênica,pode ser aplicado para problemas que requerem mais do que uma única solução. Como forma de testar esses métodos em problemas do mundo real, uma aplicação em nanotecnologia é proposta nesta tese: ao timização estrutural de fotodetectores de infravermelho de poços quânticos a partir de uma energia desejada. Os resultados apresentam novas estruturas melhores do que as conhecidas na literatura (melhoria de 59,09 por cento). / [en] Genetic programming techniques allow flexibility in the optimization process, making it possible to use them in different areas of knowledge and providing new ways for specialists to advance in their areas more quickly and more accurately.Parameter mapping approach is a numerical optimization method that uses genetic programming to find an appropriate mapping scheme among initial guesses to optimal parameters for a system. Although this approach yields good results for problems with trivial solutions, the use of large equations/trees may be required to make this mapping appropriate for more complex systems.In order to increase the flexibility and applicability of the method to systems of different levels of complexity, this thesis introduces a generalization by thus using multi-gene genetic programming to perform a multivariate mapping, avoiding large complex structures.Three sets of benchmark functions, varying in complexity and dimensionality, were considered. Statistical analyses carried out suggest that this new method is more flexible and performs better on average, considering challenging benchmark functions of increasing dimensionality.This thesis also presents an improvement of this new method for multimodal numerical optimization.This second algorithm uses some niching techniques based on the clearing procedure to maintain the population diversity. A multimodal benchmark set with different characteristics and difficulty levels to evaluate this new algorithm is used. Statistical analysis suggested that this new multimodal method using multi-gene genetic programming can be used for problems that requires more than a single solution. As a way of testing real-world problems for these methods, one application in nanotechnology is proposed in this thesis: the structural optimization of quantum well infrared photodetector from a desired energy.The results present new structures better than those known in the literature with improvement of 59.09 percent.