M.Sc. / Determining the identity of a person has become vital in today’s world. Emphasis on security has become increasingly more common in the last few decades, not only in Information Technology, but across all industries. One of the main principles of security is that a system only be accessed by a legitimate user. According to the ISO 7498/2 document [1] (an international standard which defines an information security system architecture) there are 5 pillars of information security. These are Identification/Authentication, Confidentiality, Authorization, Integrity and Non Repudiation. The very first line of security in a system is identifying and authenticating a user. This ensures that the user is who he/she claims to be, and allows only authorized individuals to access your system. Technologies have been developed that can automatically recognize a person by his unique physical features. This technology, referred to as ‘biometrics’, allows us to quickly, securely and conveniently identify an individual. Biometrics solutions have already been deployed worldwide, and it is rapidly becoming an acceptable method of identification in the eye of the public. As useful and advanced as unimodal (single biometric sample) biometric technologies are, they have their limits. Some of them aren’t completely accurate; others aren’t as secure and can be easily bypassed. Recently it has been reported to the congress of the U.S.A [2] that about 2 percent of the population in their country do not have a clear enough fingerprint for biometric use, and therefore cannot use their fingerprints for enrollment or verification. This same report recommends using a biometric system with dual (multimodal) biometric inputs, especially for large scale systems, such as airports. In this dissertation we will investigate and compare multimodal biometric techniques, in order to determine how much of an advantage lies in using this technology, over its unimodal equivalent.

This thesis consists of three papers on matching and weighting methods for causal inference. The first paper conducts a Monte Carlo simulation study to evaluate the performance of multivariate matching methods that select a subset of treatment and control observations. The matching methods studied are the widely used nearest neighbor matching with propensity score calipers, and the more recently proposed methods, optimal matching of an optimally chosen subset and optimal cardinality matching. The main findings are: (i) covariate balance, as measured by differences in means, variance ratios, Kolmogorov-Smirnov distances, and cross-match test statistics, is better with cardinality matching since by construction it satisfies balance requirements; (ii) for given levels of covariate balance, the matched samples are larger with cardinality matching than with the other methods; (iii) in terms of covariate distances, optimal subset matching performs best; (iv) treatment effect estimates from cardinality matching have lower RMSEs, provided strong requirements for balance, specifically, fine balance, or strength-k balance, plus close mean balance. In standard practice, a matched sample is considered to be balanced if the absolute differences in means of the covariates across treatment groups are smaller than 0.1 standard deviations. However, the simulation results suggest that stronger forms of balance should be pursued in order to remove systematic biases due to observed covariates when a difference in means treatment effect estimator is used. In particular, if the true outcome model is additive then marginal distributions should be balanced, and if the true outcome model is additive with interactions then low-dimensional joints should be balanced. The second paper focuses on longitudinal studies, where marginal structural models (MSMs) are widely used to estimate the effect of time-dependent treatments in the presence of time-dependent confounders. Under a sequential ignorability assumption, MSMs yield unbiased treatment effect estimates by weighting each observation by the inverse of the probability of their observed treatment sequence given their history of observed covariates. However, these probabilities are typically estimated by fitting a propensity score model, and the resulting weights can fail to adjust for observed covariates due to model misspecification. Also, these weights tend to yield very unstable estimates if the predicted probabilities of treatment are very close to zero, which is often the case in practice. To address both of these problems, instead of modeling the probabilities of treatment, a design-based approach is taken and weights of minimum variance that adjust for the covariates across all possible treatment histories are directly found. For this, the role of weighting in longitudinal studies of treatment effects is analyzed, and a convex optimization problem that can be solved efficiently is defined. Unlike standard methods, this approach makes evident to the investigator the limitations imposed by the data when estimating causal effects without extrapolating. A simulation study shows that this approach outperforms standard methods, providing less biased and more precise estimates of time-varying treatment effects in a variety of settings. The proposed method is used on Chilean educational data to estimate the cumulative effect of attending a private subsidized school, as opposed to a public school, on students’ university admission tests scores. The third paper is centered on observational studies with multi-valued treatments. Generalizing methods for matching and stratifying to accommodate multi-valued treatments has proven to be a complex task. A natural way to address confounding in this case is by weighting the observations, typically by the inverse probability of treatment weights (IPTW). As in the MSMs case, these weights can be highly variable and produce unstable estimates due to extreme weights. In addition, model misspecification, small sample sizes, and truncation of extreme weights can cause the weights to fail to adjust appropriately for observed confounders. The conditions the weights need to satisfy in order to provide close to unbiased treatment effect estimates with a reduced variability are determined and the convex optimization problem that can be solved in polynomial time to obtain them is defined. A simulation study with different settings is conducted to compare the proposed weighting scheme to IPTW, including generalized propensity score estimation methods that also consider explicitly the covariate balance problem in the probability estimation process. The applicability of the methods to continuous treatments is also tested. The results show that directly targeting balance with the weights, instead of focusing on estimating treatment assignment probabilities, provides the best results in terms of bias and root mean square error of the treatment effect estimator. The effects of the intensity level of the 2010 Chilean earthquake on posttraumatic stress disorder are estimated using the proposed methodology.

Thesis (M.S.)--George Mason University, 2008. / Vita: p. 29. Thesis director: Jennifer Weller. Submitted in partial fulfillment of the requirements for the degree of Master of Science in Bioinformatics. Title from PDF t.p. (viewed July 7, 2008). Includes bibliographical references (p. 28). Also issued in print.

This thesis focuses on two central classes of problems in discrete optimization: matching and scheduling. Matching problems lie at the intersection of different areas of mathematics, computer science, and economics. In two-sided markets, Gale and Shapley's model has been widely used and generalized to assign, e.g., students to schools and interns to hospitals. The goal is to find a matching that respects a certain concept of fairness called stability. This model has been generalized in many ways. Relaxing the stability condition to popularity allows to overcome one of the main drawbacks of stable matchings: the fact that two individuals (a blocking pair) can prevent the matching from being much larger. The first part of this thesis is devoted to understanding the complexity of various problems around popular matchings. We first investigate maximum weighted popular matching problems. In particular, we show various NP-hardness results, while on the other hand prove that a popular matching of maximum weight (if any) can be found in polynomial time if the input graph has bounded treewidth. We also investigate algorithmic questions on the relationship between popular, stable, and Pareto optimal matchings. The last part of the thesis deals with a combinatorial scheduling problem arising in cyber-security. Moving target defense strategies allow to mitigate cyber attacks. We analyze a strategic game, PLADD, which is an abstract model for these strategies.

This thesis consists of three parts, each of independent interest, yet tied together by the problem of matching with mismatches. In the first chapter, we present a motivated exposition of a new randomized algorithm for indexed matching with mismatches which, for constant error (substitution) rates, locates a substring of length m within a string of length n faster than existing algorithms by a factor of O(m/ log(n)). The second chapter turns from this theoretical problem to an entirely practical concern: delta compression of executable code. In contrast to earlier work which has either generated very large deltas when applied to executable code, or has generated small deltas by utilizing platform and processor-specific knowledge, we present a naïve approach — that is, one which does not rely upon any external knowledge — which nevertheless constructs deltas of size comparable to those produced by a platformspecific approach. In the course of this construction, we utilize the result from the first chapter, although it is of primary utility only when producing deltas between very similar executables. The third chapter lies between the horn and ivory gates, being both highly interesting from a theoretical viewpoint and of great practical value. Using the algorithm for matching with mismatches from the first chapter, combined with error correcting codes, we give a practical algorithm for “universal” delta compression (often called “feedback-free file synchronization”) which can operate in the presence of multiple indels and a large number of substitutions.

Project (M.S.)--Brigham Young University. Dept. of Statistics, 2006. / Includes bibliographical references (p. 45-46).

Thesis (M.S.)--West Virginia University, 2004. / Title from document title page. Document formatted into pages; contains vii, 64 p. : ill. (some col.). Includes abstract. Includes bibliographical references (p. 63-64).

Internet use is considered a lever for empowering patients, levelling inequalities and reducing healthcare expenditure. However, with digital inclusion, health provision quality and health system efficiency high on the UK and EU policy agendas, we need to better understand the relationship between Internet use and health outcomes to assess potential benefits and adverse effects. This research addresses the question of how Internet use influences individuals' health service use and their perceived health in the context of England. Focusing on health information-seeking, it analyses variations across different kinds of users, mechanisms between Internet use and both health outcomes, and the role of individual and contextual factors in this relationship. To answer this question, this research uses a mixed methods approach. Quantitative data from the Oxford Internet Surveys (OxIS), the English census and Hospital Episode Statistics (HES) was connected through spatial microsimulation based on output areas. Qualitative data was collected through semi-structured, face-to-face interviews, primarily with former OxIS participants from output areas in the quantitative strand. The quantitative data was revisited based on emerging interview themes. The results indicate that Internet use influences perceived health and health service use via various mechanisms based on the Internet's content, mediation and connection affordances. However, the boundaries between users and non-users are blurry and outcomes vary for different types of individuals, classified here as learners, pragmatists, sceptics, worriers, delegators and adigitals. Age, education, socioeconomic status, long-term health conditions, and geographic context influence Internet use and health outcomes separately, while the social context shapes their relationship too. These findings advance the theoretical understanding of Internet-based health outcomes, and provide practical implications for health professionals and policymakers with insights down to the local level. Moreover, this research demonstrates how novel insights for public wellbeing can be derived from qualitatively enriched secondary data in privacy-preserving and cost-effective ways.

Within the DIECOFIS framework, ec3, the Division of Business Statistics from the Vienna University of Economics and Business Administration and ISTAT worked together to find methods to create a comprehensive database of enterprise data required for taxation microsimulations via integration of existing disparate enterprise data sources. This paper provides an overview of the broad spectrum of investigated methodology (including exact and statistical matching as well as imputation) and related statistical quality indicators, and emphasises the relevance of data integration, especially for official statistics, as a means of using available information more efficiently and improving the quality of a statistical agency's products. Finally, an outlook on an empirical study comparing different exact matching procedures in the maintenance of Statistics Austria's Business Register is presented.

Ziel dieser Arbeit ist es, robuste und performante Algorithmen für die Fusion von polizeilichen Unfalldaten zur Testszenariengenerierung im Rahmen der Absicherung automatisierter Fahrfunktionen zu generieren. In dieser Arbeit werden dabei Methoden zur Datenfusion in Kombination mit generativen und Klassifikationsmodellen untersucht. Eine spezifische Variable vom Empfänger wird während des Datenfusionsverfahrens im Voraus entfernt. Ein Spender mit den gemeinsamen Variablen wird verwendet, um die Vorhersage für die fehlende spezifische Variable im Empfänger zu erhalten. Als Methode werden Ensembles aus Distance-Hot-Deck und Machine-Learning Verfahren für die Vorhersage verwendet. Nach der Vorhersage werden die Ergebnisse anhand ausgewählter Bewertungsmetriken bewertet. Darüber hinaus werden zwei generative Modelle eingeführt, um Datensätze unterschiedlicher Qualität zu synthetisieren. Ziel ist es, die Robustheit der Ensembles mit den synthetisierten „Rauschdaten“ zu testen und die Performance von Ensembles mit den synthetisierten Daten hoher Qualität zu verbessern. Schließlich können Erkenntnisse darüber gewonnen werden, welche Ensembles die besten Ergebnisse für die Datenfusion liefern.:1. Einleitung 2. Grundlagen 3. Randbedingungen 4. Vorgehensweise 5. Ergebnisse 6. Diskussion & Ausblick