University of Maryland, Baltimore County
Not a member yet
17643 research outputs found
Sort by
Entropy rate estimators and their application to blind source separation
No full textThe characterization of a signal is one of the key issues in statistical signal processing and is vital to the success of many methods. Thus, the goal of characterization techniques is to achieve a desirable balance between flexibility and maintaining a simple form, so as to allow for generalization and efficient implementation. In this dissertation, we propose a new density estimation technique, which uses the principle of maximum entropy with Gaussian kernels. By jointly using global and local measuring functions, the new density estimation method enjoys a high level of flexibility while providing a simple parametric form for the estimated density function. When considering a signal with sample dependence, the signal cannot be fully characterized by its density function. To address this issue, we propose the use of entropy rate to provide a powerful tool for data analysis, since it accounts for all types of diversity: higher-order statistics, sample dependence, noncircularity for a complex-valued process, and dependence across processes for a vector-valued process. We propose entropy rate estimators for real-valued processes based on both Markovian and invertible source models. We extend the entropy rate estimator to complex-valued and vector-valued processes by using a widely linear whitening filter and vector autoregressive model, respectively. The new entropy rate estimators are shown to exploit all these statistical properties effectively. These entropy rate estimators can be used in a wide array of applications in statistical signal processing. In this work, we mainly focus on: model order selection, real-valued independent component analysis, complex-valued independent component analysis, and independent vector analysis. Detecting the number of latent signals in a given number of observations, or order detection, is a key issue in many signal processing problems. Many of the commonly used methods to solve this problem are based on information theoretic criteria. In order to facilitate the use of information theoretic criteria, these approaches address the issue of dependent samples by downsampling the dataset. However, downsampling the data decreases the sample size, which can cause degradation in the accuracy of the order estimation. Using an entropy rate estimator, we write the likelihood based on the entire data set and use it in an information theoretic framework. We propose two entropy-rate-based order estimation methods and show the performance improvement they offer with examples. In addition, the consistency of the entropy-rate-based methods is shown when the sample size goes to infinity. By assuming latent sources are statistically independent, independent component analysis separates underlying sources from a given linear mixture. Since in many applications, latent sources are both non-Gaussian and have sample dependence, it is desirable to exploit both properties jointly. In this work, we propose to use the mutual information rate to construct a general framework for analysis and derivation of algorithms that account for both types of diversity. We introduce four algorithms for the problem of independent component analysis, and show that their performance can approach the Cramer-Rao lower bound. Complex-valued signals frequently arise in applications such as communications, radar, and biomedicine, as well as processing when real-valued signals are in a transform domain. Independent component analysis, when implemented in the complex domain, enjoys the addition of yet another type of diversity, noncircularity of the sources---underlying components. Using the new entropy rate estimator, we develop a new complex-valued independent component analysis algorithm, which exploits all three types of diversity, non-Gaussianity, sample dependence, and noncircularity. In addition, we establish the general form of the Fisher information matrix and Cramer-Rao lower bound and show the identification conditions for the complex-valued independent component analysis problem. Experimental results show that the new method accounts for all three types of diversity effectively and provides desirable performance for a wide variety of sources. Additionally, the entropy rate estimator for vector-valued processes is used for the independent vector analysis problem, which is a generalization of independent component analysis to multiple datasets and adds the use of one more diversity, the dependence across multiple datasets. Using mutual information rate, we propose a new method that takes all these types of diversity into account
The Relation of Motivation and Self-efficacy to Consistency of Condom Use: The Role of Sex Under the Influence of Alcohol
No full textYouth ages 15 to 24 account for half of the estimated 20 million new cases of Sexually Transmitted Infections (STIs) that occur each year (CDC, 2013). Consistent condom use is an important intervention target to prevent STI contraction among youth. The present study evaluated whether motivation and self-efficacy to use condoms predict subsequent consistency of condom use in a community sample (N=166) of female youth (ages 13-21) at increased risk for STI contraction. Additionally, the current study investigated if these relations are moderated by frequency of sex under the influence. Exploratory analyses investigated these relations in sex events with steady boyfriends only separately from sex events with other guys only. Data were analyzed from Project SAFE, a randomized controlled clinical trial investigating the efficacy of a Computer-Assisted Motivational Intervention compared to Didactic Educational Counseling for sexual risk and protective behaviors. Both motivation and self-efficacy to use condoms demonstrated a significant positive linear relation with subsequent consistency of condom use. However, there was limited evidence for the moderating effect of frequency of sex under the influence. There was some indication that the pattern of results differed in exploratory analyses including only sex events with steady boyfriends compared to sex events with other guys. Analyses were re-estimated including past consistency of condom use as a covariate. In these analyses there was some evidence for a negative main effect of frequency of sex under the influence. These finding have implications for the development of tailored interventions and future research
BLACK FACES IN BLUE UNIFORMS: DESEGREGATING THE BALTIMORE POLICE DEPARTMENT 1920-1950
No full textIn 1937, the Baltimore city police department hired its first African-American officer. Her hire and the subsequent hire, seven months later, of four black male officers represented the culmination of nearly two decades of civil rights activism by a coalition of print media, middle-class reformers, and cooperative politicians. I argue that civil rights activity in Baltimore, especially in the first half of the twentieth century, was generally nonviolent and led by a strong-willed, generally secular cadre of middle-class African Americans. In addition, because of its border state status, racial gains occurred earlier in Baltimore than they did in Dixie, but not as early as they did in more northern cities. The leaders of Baltimore's civil rights movement used their political and journalistic influence to call for and effect change. The process of desegregating the city's police department is a clear example of this. I begin with a general narrative of the campaign for desegregation between 1920 and 1950. Using the reporting and editorializing of the Baltimore Afro-American, I explain the importance of desegregation to the community, not only as a tangible example of racial progress, but as acknowledgment of the rising civic power of African-American Baltimoreans. I explore issues related to the police department and its integration, including the racial and sexual power dynamics inherent between any police force and its community, especially in the United States in the nineteenth and twentieth centuries. I also examine the economic class of the activists and the first few officers as a means of reiterating the fundamentally middle-class nature of this particular civil rights campaign
Structural Basis of HIV-1 Gag Membrane Targeting by The N-Terminal Matrix Domain
No full textAssembly of HIV-1 particles is initiated by the trafficking of viral Gag polyproteins from the cytoplasm to the plasma membrane (PM), where they co-localize and bud to form immature particles. Membrane targeting is mediated by the N-terminally myristylated matrix (MA) domain of Gag and is dependent on the PM marker phosphatidylinositol-4,5-bisphosphate [PI(4,5)P2]. Conversely, the mechanism of trafficking Gag molecules to specific assembly sites is unclear. In addition to PI(4,5)P2 dependent targeting, cellular proteins TIP47 and AP-3δ have been implicated to bind MA for Gag trafficking to assembly sites. In an effort to investigate the structural basis of the proposed TIP47 and AP-3δ trafficking of MA, we conducted NMR titration experiments of recombinant TIP47 and AP-3δ with MA. Our data provide new information demonstrating that (1) AP-3δ does not interact with MA and (2) Tip47 binds MA, however, was not necessary for efficient Viral Env incorporation during viral assembly in vivo. Previous in vitro membrane binding assays revealed that PI(4,5)P2 can influence membrane affinity in a concentration dependent manner. Structural studies using a water soluble form of PI(4,5)P2 [tr-PI(4,5)P2] revealed that tr-PI(4,5)P2 binds directly to MA with the 2'acyl chain unexpectedly interacting with a hydrophobic cleft on MA. This resulted in a proposed extended lipid-binding model for MA binding at the PM. Here we report that tr-PI(4,5)P2 binds non-discriminately to hydrophobic surface patches of proteins that do not bind membranes, including HIV-1 capsid (CA), which prompted us to examine MA interactions with native PI(4,5)P2 and other membrane constituents in the presence of membrane mimetic liposomes and bicelles. NMR titration experiments with detergent bicelles revealed a small subset of surface and myristate-associated residues that are sensitive to the presence of PI(4,5)P2, whereas cleft residues that bound the 2?-chains of tr-PI(4,5)P2 molecules in the absence of bicelles or liposomes were unperturbed by the addition of PI(4,5)P2. Our findings call to question extended-lipid MA:membrane binding models and suggest a targeting mechanism that includes a role for liquid disordered regions of the PM
RECONSTRUCTING THE EVOLUTIONARY HISTORY OF DARTERS (PERCIDAE: ETHEOSTOMATINAE): EPIGENETIC DIVERGENCE AS A FIRST STEP IN SPECIATION
No full textThis doctoral research reconstructs the evolutionary history and investigates the possibility of changes in gene expression, rather than nucleotide sequence, being a first step in speciation in the second most speciose group of freshwater fishes in North America. To gain a better understanding of the incredible diversity in this group of fishes known as darters (Percidae: Etheostomatinae), I first determine evolutionary relationships among 69 and then 101 different species representing four genera of darters using a genome-scan approach known as AFLP. Prior to this work, AFLPs had only been used for recent divergences, with the oldest reconstruction dating to approximately 15 million years. In silico studies had predicted that increased homoplasy would weaken the utility of AFLPs in reconstructing ancient evolutionary relationships. We found that dense taxon sampling increased accuracy and resolution in this group of fishes. We were able to resolve problematic nodes and reconstruct the evolutionary history of darters with moderate to strong statistical support at ancient divergences (>25 mya) and thus champion this method for deeper relationships. In addition, we were able to propose a new taxonomic classification of darters based on our and other recent phylogenies. Phylogenetic reconstruction revealed a large amount of interspecific genetic diversity in this group of fishes. This finding led us to question whether fast-evolving changes in gene expression via changes to the methylome, rather than nucleotide sequence, could initiate speciation in this group of fishes. We investigated whether epigenetic divergence acts as a first step in speciation in Etheostoma, a genus of darters. Using Methyl-Sensitive Amplified Polymorphisms (MSAP), we analyzed population structure and differentiation in the methylome and genome across 5 populations of E. olmstedi (the tessellated darter) in the Patuxent River and 6 populations in the Potomac River in Maryland, USA. We predicted that epigenetic variation would be greater than genetic variation both within and between populations, with populations showing greater differentiation in their methylomes compared to their genomes. We also investigated the relationship between genetic divergence, epigenetic divergence, and the strength of various reproductive barriers across more distantly related, geographically isolated species in the subgenera Ulocentra, Etheostoma and Nanostoma. Since previous studies have shown behavioral isolation to be the strongest reproductive barrier between darter species, we predicted that behavioral isolation would be better predicted by epigenetic as compared to genetic differences. Our results indicate that the methylome is indeed changing faster than the nucleotide sequence across all levels of divergence, and that epigenetic divergence better predicts behavioral reproductive isolation than genetic divergence, supporting the hypothesis that epigenetic divergence can act as a first step in speciation in this group
Acculturation, construction of gender and social identity in a sample of Afghan immigrants in the U.S.
No full textThis qualitative study, informed by the phenomenological approach, was an effort to examine the post-migration experiences of Afghan immigrants in the United States. Sarbin's social role theory framed the exploration of three processes: acculturation, construction of social identity and gender. Findings reveal that the three processes were not entirely distinct but influenced each other in important ways. Additionally, the pre-migration and post-migration factors shaped how they adapted within each of the four ecologies defined by Sarbin (1970). Furthermore, the sociopolitical context of relations between the United States and Afghanistan shaped the processes of acculturation, construction of gender and social identity of the participants. The implications of these findings for future research and programs are discussed
THE HISTORY OF KINDERGARTEN IN BALTIMORE CITY, MARYLAND
No full textABSTRACT History Department. Kindergarten was originally designed in 1837 to create social interaction and learning through play, became a sculpting tool for political issues at the turn of the twentieth century. This thesis explores how the changing racial and ethnic demographic of students in Baltimore affected policies related to kindergarten in the city. It shows kindergarten as a tool for social activism under the guise of teaching American patriotism. This thesis uses government legislation and secondary sources to show how education was influenced by increased immigration and segregation in Baltimore City during the mid-twentieth century. The significance of this thesis lies in its exposure of class, ethnic, and racial stratification in education at its most basic level; class in this regard refers to socioeconomic status influenced by ethnicity and race. Kindergarten, while being promoted as a way to blend cultures in a changing society, actually reinforced the disparity among classes by allowing elites to capitalize on their fears related to these differing cultures
Performance Evaluation of Probabilistic Latent Semantic Analysis for Unstructured Social Media Data
No full textBig data analytics is being applied in many fields today to mine unstructured data such as social media blogs or medical records. We focus this thesis on two popular analysis techniques, the methods of Latent Semantic Analysis(LSA) and Probabilistic Latent Semantic Analysis(PLSA), both used for interpreting or extracting concepts and relationships from data. As a use case, we propose to compare their performances in identifying communities from Twitter data sets during natural disasters such as Hurricanes. Latent semantic analysis uses statistical computations, typically singular value decomposition, to find semantic or contextual meaning from the data. It finds relationships between terms and concepts in an unstructured data set. Probabilistic latent semantic analysis or indexing is another method based on Bayesian analysis that typically is used for two-mode data. The objective is to compare these two methods on a large set of social media documents related to Hurricane Sandy in order to form clusters of similar concepts. We then compare the performance of these two methods to determine their relative performance in determining communities and hidden topics, e.g. finding clusters of similar topics like power outages, floods, gas outages, etc. We apply two clustering methods, K-Means and Affinity Propagation to form clusters in the data. Finally, we present the results by applying external methods of evaluation after creating a test data-set to compare the performance of these two methods. Metrics like Precision, Recall, confusion matrix are used to evaluate the performance of our system. The evaluation showed us that in almost all the scenarios, PLSA works better than LSA in finding out hidden relationships and structures. Whereas LSA is slightly faster than PLSA
USERS' PRIVACY AND SECURITY BEHAVIORS ON MOBILE DEVICES.
No full textPreferences and behaviors for privacy management with mobile applications are difficult to capture. Previous measures are mostly based on self-report data, which often does not accurately predict actual user behavior. A deeper understanding was sought, gleaned from observing actual practices. This thesis analyzes 11,777 applications from the Google Play marketplace in order to determine the impact of privacy settings on purchase behavior. This was done by looking at the effect of the number of privacy concessions as well as the effect of individual concessions and category on number of downloads. It was found that users of paid applications do not have a preference for fewer privacy concessions. This study further reinforces the disconnect between the user's often stated preference for privacy and their actual behavior – a discrepancy known as the “privacy paradox ”. Theoretical and practical implications are discussed
The Visual Ecology of Stomatopod Larvae
No full textDespite researchers' interest in adult stomatopod vision, only a handful of studies have investigated stomatopod larval visual ecology, the function of the larval compound eye, or its transition into the adult eye structures. In order to address this knowledge deficit, I tested several hypotheses surrounding larval eye physiology and ontogeny using microspectrophotometry (MSP) and electroretinography (ERG). MSP was used to investigate visual pigment diversity in the retinas of different species of stomatopod larvae. Together with data previously published, these data provide further support that larvae possess a single spectral class of photoreceptor. Surprisingly, the peak spectral absorbance varied significantly among sympatric species, with four species falling into a short-wavelength and the remainder into a long-wavelength sensitivity class. These results may be explained by differences in behavioral ecology or by different developmental stages of individual test subjects. To investigate ontogenetic changes in retinal function, ERG intensity responses were measured from different stage retinas of a single species, Squilla empusa. Initial MSP data from transitioning retinas found no evidence of visual pigment absorption in newly emerged adult retinas, implying a putative delay in photoreceptive function. Robust ERG responses to light stimuli, however, were observed from the earliest presence of emerging adult retinas, rejecting this hypothesis. These data also suggest an increase in the response dynamics of stomatopod retinas with ontogeny as well as an increase in irradiance sensitivity during the double-retina phase. To investigate the visual ecology of stomatopod larval light-reflecting structures, I characterized the function and structure of eyeshine overlying the retina. In situ photography of larvae eyeshine demonstrated its function as a contrast reducer, or camouflage. Calculations of eyeshine reflectance spectra in their natural setting revealed novel, spectral matching with the background environment. Eyeshine structures were characterized via transmission electron microscopy (TEM) and identified as amorphous, three-dimensional photonic crystal arrays of spherical vesicles that coherently scatter light. The photonic mechanism of eyeshine production was tested using a modified Bragg-theory model and dimensions from TEM micrographs. These eyeshine data as well as data regarding the physiology of stomatopod larval vision provide a strong foundation for future investigations into the ontogeny of vision in stomatopod crustaceans