A new adaptive learning Artificial Immune System (AIS) based committee machine is developed in this thesis. The new proposed approach efficiently tackles the general problem of clustering high-dimensional data. In addition, it helps on deriving useful decision and results related to other application domains such classification and prediction. Artificial Immune System (AIS) is a branch of computational intelligence field inspired by the biological immune system, and has gained increasing interest among researchers in the development of immune-based models and techniques to solve diverse complex computational or engineering problems. This work presents some applications of AIS techniques to health problems, and a thorough survey of existing AIS models and algorithms. The main focus of this research is devoted to building an ensemble model integrating different AIS techniques (i.e. Artificial Immune Networks, Clonal Selection, and Negative Selection) for classification applications to achieve better classification results. A new AIS-based ensemble architecture with adaptive learning features is proposed by integrating different learning and adaptation techniques to overcome individual limitations and to achieve synergetic effects through the combination of these techniques. Various techniques related to the design and enhancements of the new adaptive learning architecture are studied, including a neuro-fuzzy based detector and an optimizer using particle swarm optimization method to achieve enhanced classification performance. An evaluation study was conducted to show the performance of the new proposed adaptive learning ensemble and to compare it to alternative combining techniques. Several experiments are presented using different medical datasets for the classification problem and findings and outcomes are discussed. The new adaptive learning architecture improves the accuracy of the ensemble. Moreover, there is an improvement over the existing aggregation techniques. The outcomes, assumptions and limitations of the proposed methods with its implications for further research in this area draw this research to its conclusion.

This research attempts to confirm the reliability and construct validity of a personnel selection instrument called a Situational Judgment Test (SJT) through reliability analysis and factor analysis. The existing literature on SJTs is reviewed, including the advantages of using SJTs in personnel selection as well as the debate on whether SJTs measure a single construct or whether they can be multidimensional depending on the content. The specific SJT in this research was theoretically developed and received expert ratings to assess four general constructs: problem solving, planning, priority setting, and leadership. No support from alpha internal consistency reliability analysis was found for the assembly of these items into the four a priori subscales, thus assembly of these items into the theoretical subscales and scales was not supported.

The study investigated the predictive validity of the California Psychological Inventory (CPI) as a tool for sales selection. Two analyses were conducted. Study 1 consisted of 20 male home improvement representatives. The average net and gross closing ratios for a six month period were used as the criteria. The results indicted that none of the CPI scales differentiated between poor and good performers when correlated with the secondary criterion of gross closing ratios. These findings were contrary to a previous concurrent validity study. Study 2 investigated month to month retention/separation for 61 home improvement representatives, to determine if the CPI differentiated between short and long term success. Phi coefficients showed no statistical significance between retention/separation and the CPI profile score over time.

There has been extensive research done in the area of Semiparametric Regression. These techniques deliver substantial improvements over previously developed methods, such as Ordinary Least Squares and Kernel Regression. Two of these hybrid techniques: Model Robust Regression 1 (MRR1) and Model Robust Regression 2 (MRR2) require the choice of an appropriate bandwidth for smoothing and a mixing parameter that allows a portion of a nonparametric fit to be used in fitting a model that may be misspecifed by other regression methods. The current method of choosing the bandwidth and mixing parameter does not guarantee the optimal choices in either case. The immediate objective of the current work is to address this process of choosing the optimal bandwidth and mixing parameter and to examine the behavior of these estimates using 3D plots. The 3D plots allow us to examine how the semiparametric techniques: MRR1 and MRR2, behave for the optimal (AVEMSE) selection process when compared to data-driven selectors, such as PRESS* and PRESS**. It was found that the structure of MRR2 behaved consistently under all conditions. MRR2 displayed a wider range of "acceptable" values for the choice of bandwidth as opposed to a much more limited choice when using MRR1. These results provide general support for earlier fndings by Mays et al. (2000).

As Web services grow in maturity and use, so do the methods which are being used to test and maintain them. Regression Testing is a major component of most major testing systems but has only begun to be applied to Web services. The majority of the tools and techniques applying regression test to Web services are focused on test-case generation, thus ignoring the potential savings of regression test selection. Regression test selection optimizes the regression testing process by selecting a subset of all tests, while still maintaining some level of confidence about the system performing no worse than the unmodified system. A safe regression test selection technique implies that after selection, the level of confidence is as high as it would be if no tests were removed. Since safe regression test selection techniques generally involve code-based (white-box) testing, they cannot be directly applied to Web services due to their loosely-coupled, standards-based, and distributed nature. A framework which automates both the regression test selection and regression testing processes for Web services in a decentralized, end-to-end manner is proposed. As part of this approach, special consideration is given to the concurrency issues which may occur in an autonomous and decentralized system. The resulting synchronization method will be presented along with a set of algorithms which manage the regression testing and regression test selection processes throughout the system. A set of empirical results demonstrate the feasibility and benefit of the approach.

<p>Fitting statistical models is computationally challenging when the sample size or the dimension of the dataset is huge. An attractive approach for down-scaling the problem size is to first partition the dataset into subsets and then fit using distributed algorithms. The dataset can be partitioned either horizontally (in the sample space) or vertically (in the feature space), and the challenge arise in defining an algorithm with low communication, theoretical guarantees and excellent practical performance in general settings. For sample space partitioning, I propose a MEdian Selection Subset AGgregation Estimator ({\em message}) algorithm for solving these issues. The algorithm applies feature selection in parallel for each subset using regularized regression or Bayesian variable selection method, calculates the `median' feature inclusion index, estimates coefficients for the selected features in parallel for each subset, and then averages these estimates. The algorithm is simple, involves very minimal communication, scales efficiently in sample size, and has theoretical guarantees. I provide extensive experiments to show excellent performance in feature selection, estimation, prediction, and computation time relative to usual competitors.</p><p>While sample space partitioning is useful in handling datasets with large sample size, feature space partitioning is more effective when the data dimension is high. Existing methods for partitioning features, however, are either vulnerable to high correlations or inefficient in reducing the model dimension. In the thesis, I propose a new embarrassingly parallel framework named {\em DECO} for distributed variable selection and parameter estimation. In {\em DECO}, variables are first partitioned and allocated to m distributed workers. The decorrelated subset data within each worker are then fitted via any algorithm designed for high-dimensional problems. We show that by incorporating the decorrelation step, DECO can achieve consistent variable selection and parameter estimation on each subset with (almost) no assumptions. In addition, the convergence rate is nearly minimax optimal for both sparse and weakly sparse models and does NOT depend on the partition number m. Extensive numerical experiments are provided to illustrate the performance of the new framework.</p><p>For datasets with both large sample sizes and high dimensionality, I propose a new "divided-and-conquer" framework {\em DEME} (DECO-message) by leveraging both the {\em DECO} and the {\em message} algorithm. The new framework first partitions the dataset in the sample space into row cubes using {\em message} and then partition the feature space of the cubes using {\em DECO}. This procedure is equivalent to partitioning the original data matrix into multiple small blocks, each with a feasible size that can be stored and fitted in a computer in parallel. The results are then synthezied via the {\em DECO} and {\em message} algorithm in a reverse order to produce the final output. The whole framework is extremely scalable.</p> / Dissertation

Recruitment and selection of employees is one of the most important activities in the personal department of company. The aim of my thesis is to analyze the process of recruitment and selection of employees, firstly from the theoretical point of view and further from the practical experience in the company ČSOB, a.s. I have summarized theoretical findings and recommendations regarding the process of recruitment and selection of employees according to the available literature in the practical part of my thesis. The practical part continue with the analysis of this process in the company ČSOB, a.s. This analysis has been realized on the basis of internal documentation, interviews with employees of the personal department and my own findings acquired by participation in selection procedure and assessment centre. As a last source of information for evaluation of this process I have used results of questionnaire examination that was realized in cooperation with the bank in April 2010. To finish my thesis I have summarized possible recommendations for improvement of the process of acquirement and selection of employees in ČSOB, a.s.

We present a software that can dynamically determine what machine learning algorithm is best to use in a certain situation given predefined traits. The produced software uses ideal conditions to exemplify how such a solution could function. The software is designed to train a selection algorithm that can predict the behavior of the specified testing algorithms to derive which among them is the best. The software is used to summarize and evaluate a collection of selection algorithm predictions to determine  which testing algorithm was the best during that entire period. The goal of this project is to provide a prediction evaluation software solution can lead towards a realistic implementation.

Thesis advisor: Philip G. Altbach / The literature on Latin American higher education indicates the existence of a relationship between socio-economic status and college enrollment. One of the hypotheses of this study was that in Haiti, socio-economic status is related not only to college access but also to students' ability to enter their preferred field of study. As a result, students from higher socio-economic status were expected to report higher levels of satisfaction with their academic situation. In this quantitative survey study, an instrument was developed and administered to 742 college students in 5 different Haitian institutions in order to determine whether there exists this hypothesized relationship between students' socio-economic status and their satisfaction with their academic situation. Data analysis revealed a weak, negative relationship between students' socio-economic status and their satisfaction with their academic situation. No significant relationship could be established between socio-economic status and access to a preferred field of study, across all students. Instead the study found what seems to be a paradox: although a majority of students were not able to access their desired field of study, they showed a high level of satisfaction with their academic situation. This paradox is explained by the importance of intrinsic factors as well as job prospect in predicting students' satisfaction. Other findings include (a) a low level of participation for women in Haitian higher education, (b) a lower level of satisfaction for Haitian female science, engineering, and technology students, and (c) little differentiation in academic preparation between science, engineering, and technology students and the rest of the sample. Based on the research findings, the study concludes with policy recommendations to help Haitian higher education achieve its economic development mission. / Thesis (PhD) — Boston College, 2009. / Submitted to: Boston College. Lynch School of Education. / Discipline: Educational Administration and Higher Education.

A Análise do Comportamento apresenta suas raízes nas Ciências Naturais, em especial, na Biologia. Estas raízes ofereceram uma transposição de modelos metodológicos e conceitos teóricos que foram incorporados na Análise do Comportamento. Dentre estes, encontra-se o modelo de Seleção por Consequências, proposto por B. F. Skinner. Tal modelo é baseado na teoria da Seleção Natural de Darwin. Com isto, o presente trabalho tem como objetivos analisar 1) as características da Seleção Natural e da Seleção por Consequências; 2) as aproximações e diferenças presentes na transposição proposta por Skinner; e 3) a apresentação de outros processos evolutivos concomitantes à Seleção Natural e a discussão da possibilidade e necessidade da incorporação destes. Assim, o Capítulo 1 apresenta a formulação da ideia selecionista, utilizando autores clássicos do pensamento evolutivo (Darwin e Wallace). O Capítulo 2 apresenta as características particulares do modelo de Seleção por Consequências de Skinner. Por fim, o Capítulo 3 apresenta as críticas gerais ao modelo selecionista e enuncia outros modelos evolutivos que podem ser passiveis de serem transpostos à Análise do Comportamento. Observa-se a partir das análises textuais a presença de convergências e divergências entre o modelo de Skinner e a teoria evolutiva. Ambos apresentam o ambiente como força motriz das mudanças comportamentais e evolutivas, colocando a pressão deste ambiente como consequência selecionadora das características variantes nos indivíduos. Entretanto, o modelo de Skinner apresenta disparidades e problemáticas: a) aproxima-se mais de autores como Wallace em relação à Darwin; e b) não apresenta uma atualização dos modelos disponíveis, negligenciando processos evolutivos que podem ser transpostos ao fenômeno comportamental. Ainda se discute a viabilidade da transposição, como o status teleológico do selecionismo, o caráter inédito da proposta de Skinner, e a natureza metafórica da analogia em si. Discute-se uma atualização dos conceitos por parte da Análise do Comportamento assim como a incorporação de modelos acessórios à Seleção Natural, com o objetivo de diminuir as fronteiras entre as ciências e aumentar o poder explicativo dos modelos propostos / The Behavior Analysis has its roots in the Natural Sciences, in particular in Biology. These roots offered a transposition of methodological models and theoretical concepts that have been incorporated in Behavior Analysis. Among these, there is the model of Selection by Consequences, proposed by BF Skinner. This model is based on Darwins theory of Natural Selection. Therewith, the present study aims to analyze 1) the characteristics of Natural Selection and Selection by Consequences; 2) the similarities and differences present in this transposition proposed by Skinner; and 3) the presentation of other concomitant evolutionary processes of Natural Selection and the discussion of the possibility and need to incorporate these. Thus, Chapter 1 presents the formulation of selectionist idea, using evolutionary classical authors (Darwin and Wallace). Chapter 2 presents the particular characteristics of Skinners Selection by Consequences model. Finally, Chapter 3 presents the general criticism of the selectionist model and sets out other evolutionary models that may be able to be translated at the Behavior Analysis. It is observed from the textual analyzes the presence of convergence and divergence between the model of Skinner and evolutionary theory. Both feature the environment as the driving force of the behavioral and evolutionary changes, placing the environmental pressure as the consequence that selects the characterizing variants of individuals. However, the model of Skinner presents disparities and problems: a) approaches over other authors, like Wallace, in relation to Darwin; b) does not present an update of the available models, neglecting evolutionary processes that can be transposed to the behavioral phenomenon. It has also been discussed the feasibility of implementation, as the teleological status of selectionism, the unprecedented character of Skinner\'s proposal, and the metaphorical nature of the analogy itself. It discusses need for an update of biological concepts by the Behavior Analysis field as well as the incorporation of accessories models to Natural Selection, in order to reduce the boundaries between sciences and increase the explanatory power of the proposed models