University of Maryland, Baltimore County
Not a member yet
17643 research outputs found
Sort by
History of Baltimore, 1729-1920
No full textUnfinished manuscript of a chronological history of Baltimore, Maryland, from the city's founding in 1729 to 1920. Written by Dr. Joe Arnold (1937-2004), History professor at UMBC. Chapter introductions written by Dr. Elizabeth M. Nix.Manuscript made available from a partnership between UMBC departments of History, the Center for Digital History Education, the College of Arts, Humanities, and Social Sciences, and the Albin O. Kuhn Library & Gallery. Copyediting and formatting completed 2014
Detecting Deception in Computer Mediated Communication: A Social Structural Perspective
No full textDespite the widespread use of Computer-Mediated Communication (CMC) for effective collaboration and interaction, CMC has become a growing hotbed for deception due to its provision of ubiquity, anonymity and open environment. Deception is an increasing threat to our society and to the daily communication of both individuals and groups. This dissertation aims to provide a new venue for understanding deception and for detecting deception through the identification, extraction, and application of social structural behaviors of deceptive communication. To this end, the dissertation consists of three major studies. The first study conceptualizes deception in terms of social structure by drawing on interpersonal deception theory and social network theories and proposes a research model of structural properties of deceptive communication: centrality, cohesion and similarity. Viewed from the social structural perspective, structural behaviors are denoted as the relationships between different individuals (entities) or as relatively stable patterns of relationships. The second study examines the impact of time on structural behaviors of deception based on interactional adaption theory and characteristics of temporal networks. The third study addresses the problem of automatic deception detection by extracting the structural features of deceptive communication and by combining the structural features with linguistic features. In addition, it evaluates the generality of the structural features identified from synchronous CMC to asynchronous CMC. The findings of this dissertation extend existing theories and research on explaining the effect of deception intent on structural behaviors of communication in multi-fold aspects. First, this dissertation extends the context of deception theories from interpersonal interaction to social interaction by addressing the interactional dynamics in group communication that is composed of one deceiver and multiple receivers. Second, the dissertation operationalizes the structural behaviors of deceptive CMC and extracts them from two different types of networks: static and temporal, and empirically validates the behaviors with real-world data. Third, the dissertation improves the performance of automated deception detection by incorporating the structural behaviors of deception
Using Markov Logic Networks to Provide Autonomic Management of High Performance Computing System Configurations and Operations
No full textA framework utilizing statistical relational learning methods to resolve configuration conflicts and maintain details of system state can remain functionally cohesive, domain flexible, and minimize initial configuration and state awareness in order to provide automated services for managing the complex and changing configuration and operations of High Performance Computing systems. In this dissertation, the issue of how to construct and utilize a framework built upon distributed knowledge to manage complex system configuration and operations is addressed. The focal point of the framework is the statistical relational learning model using Markov Logic Networks (MLNs), which can be applied in different ways to various types of system management problems. This framework provides a new methodology to solving these issues without relying on a centralized knowledge base. Beyond the use of MLNs, this framework contains modules covering host selection, data gathering, inference analysis, and analysis actions. To understand how the framework can be applied to various system management domains, two different areas have been selected to prove its flexibility and functionality: configuration management and job scheduling. The configuration management domain focuses on a variety of operating system-level parameters to ensure consistency and correctness. While the job scheduling domain focuses on understanding processor core temperatures and profiles across the system to minimize the thermal variance across the system using intelligent scheduling techniques. The contributions of this dissertation are: (i) Development of a framework to provide consistent system management functionality in a distributed manner using MLNs; (ii) Defining an approach to implementing this framework to solve distributed configuration management issues and analyzing these results to correct identified conflicts; and (iii) Defining a job scheduling approach for HPC systems which has a set goal of thermal balancing while utilizing this framework and analyzing its results to minimize thermal variance across a system
Youth adjustment in the context of neighborhood disadvantage: A focus on stress, coping, and mental health
No full textGrowing up in urban neighborhood disadvantage, an environment characterized by severe and pervasive stress exposure, is associated with negative outcomes for youth. African American children are disproportionately exposed to neighborhood disadvantage; yet, little research has been conducted among this population. This two-part study uses quantitative and qualitative approaches to better understand the experience of growing up in neighborhood disadvantage. The quantitative component examined the fit of a model predicting adjustment (risky behaviors; internalizing and externalizing mental health) from disadvantage (family, housing, and neighborhood systems), including the influence of age and sex. Participants included 519 African American youth, ages 12-19, from five major cities in the US, who were members of the control group of a larger study. The qualitative component explored youths' perspective on stress, coping, and mental health with particular attention to different invariant meaning structures emerging by sex, age, and neighborhood. One hundred African American youth, ages 15-23, all residents of Baltimore, were interviewed as part of a larger study. Findings from this two-part study complement each other, suggesting that African American youth growing up in disadvantaged neighborhoods are exposed to severe and chronic stress across multiple systems. The quantitative component showed that this risk exposure was associated with negative youth adjustment through direct and indirect effects. The qualitative component identified unique stressors, coping strategies, and experiences with mental health. The complexity of risk and adjustment, and their interrelatedness in the challenging environment of neighborhood disadvantage, were apparent in both components of the study. Implications for families, mental health providers, and policy makers are discussed
Developing a Cybersecurity Text Corpus and its Application for Augmenting Semantic Text Similarity
No full textThe growing use of cyber-services automatically impart great importance to cybersecurity. The Internet is a primary source of information regarding software flaws, vulnerabilities, cyber-attacks and exploits. This information is available through vulnerability databases, news articles, security bulletins and blogs. Variety of applications and security systems like Intrusion Detection Systems (IDS), Intrusion Prevention System (IPS), etc. can take advantage of this information for consolidating their infrastructure. The lack of availability of ready text corpus of high quality security information from various sources makes it difficult for these applications to use this information. To overcome this problem our work focuses on building a multi-genre corpus of security text using information retrieved from multiple internet based sources; National Vulnerabilities Database, Wikipedia articles, security blogs, security bulletins and scholarly papers. The system builds a text classifier from the initial high quality data which is used to classify and accommodate new data from these sources into the corpus. This corpus can be used by variety of applications like IDS or IPS, in variety of ways like assertion into knowledge base or extraction of named entities. Our work explores one of the applications of generating the semantic text similarity model for cybersecurity text. We use the multi-genre cybersecurity text corpus for creating the word co-occurrence model. This model can extract the synonymity between the different security terms. For example, the words 'virus' and 'malware' that have same context are scored for their degree of similarity. The word co-occurrence model is then extended to generate a semantic text similarity model.The text similarity model extracts the semantic text similarity between different security texts like titles of the papers, vulnerability descriptions, blog paragraphs, etc. The system also develops a combined text similarity model from cybersecurity similarity model and generic text similarity model. This model can be used in document mining for matching security text, clustering documents describing similar vulnerabilities and so on
Bayesian Adaptive Dose-finding Methods in Phase I Drug Combination Trials
No full textIn a single-agent dose-finding Phase I trial, the key underlying assumption is that toxicity probability increases monotonically with the dose level. However, in multi-agent trial, this assumption may not hold because the drug-drug interaction effect can either decrease or increase the joint toxicity as compared to either one used alone, which may lead to an unforeseen toxicity probability surface. In the first part of the dissertation, we develop a novel adaptive dose-finding approach which can be applied to these kinds of multi-drug combination trials under the situation of non-monotonic toxicity probability surface. In the second part of the dissertation, we extend our investigation on the drug combination dose-finding trials with late-onset toxicity outcomes and have proposed a Bayesian adaptive dose-finding design under a nonignorable missing data mechanism, and where surrogate data are available. We evaluate the operating characteristics of the aforementioned methods and also compare them with existing methods through extensive simulation studies under various scenarios. The proposed methods demonstrate satisfactory performance in general
Imperfect Learning in Video Game Artificial Intelligence
No full textVideo game AI to this day is a series of hardcoded responses to situations that may arise during the course of the game. As the game becomes more complex, the difficulty of coding such an AI increases as well. Other methods of creating a viable AI, such as neural networks, are being explored, but none are in use in commercial games today. While they are being explored, neural networks are mostly used for pre-teaching before the game goes out on the market. Neural networks that continuously learned could provide an interesting and unique experience for the player every playthrough, learning new methods each time so the game did not become stale. However, the AI could quickly become too good, making the game impossible and uninteresting. Our goal is to use the theory behind real-time learning using neural networks in games while ensuring that the player is kept engaged. To do this, we employed a GA for learning and two different methods of unlearning. This unlearning keeps the AI hovering at approximately the player's level, never becoming so good it is unkillable while still providing a challenge. The first method is a simple probability of the GA picking the worse child as opposed to the better one. The other method is an altered version of simulated annealing in which the AI never reaches a best state. Both methods worked well for adjusting to a player gradually getting better or worse, although sudden jumps in skill were more difficult to adjust to
The United States Air Force Band: Musical Ambassadors on a Cold War Stage
No full textFrom 1950 to 1960, the United States Air Force Band in Washington, DC participated in one of the most aggressive and widespread diplomatic campaigns in United States History. They completed ten international tours through Europe, the Middle East, Asia, and South America as part of the U.S. State Department's cultural diplomacy program. Conceived as a way to thwart the spread of Soviet Communism in vulnerable countries worldwide, cultural diplomacy's purpose was to promote democracy and American ideas of freedom and individuality as a way to gain global influence and contain Soviet infiltration. In researching this project, I discovered that the USAF Band exemplified U.S. cultural ambassadorship. Their performances were met with large and enthusiastic audiences. Through their musical mission, the band cultivated some extraordinary relationships within some very tenuous environments, strengthening US goodwill efforts abroad and perpetuating democratic ideals.  
Development Of Advanced Sandwich Core Topologies Using Fused Deposition Modeling And Electroforming Processes
No full textNew weight efficient materials are needed to enhance the performance of vehicle systems allowing increased speed, maneuverability and fuel economy. This work leveraged a multi-length-scale composite approach combined with hybrid material methodology to create new state-of-the-art additive manufactured sandwich core material. The goal of the research was to generate a new material to expands material space for strength versus density. Fused-Deposition-Modeling (FDM) was used to remove geometric manufacturing constraints, and electrodepositing was used to generate a high specific-strength, bio-inspired hybrid material. Microtension samples (3mm x 1mm with 250?m x 250?m gage) were used to investigate the electrodeposited coatings in the transverse (TD) and growth (GD) directions. Three bath chemistries were tested: copper, traditional nickel sulfamate (TNS) nickel, and nickel deposited with a platinum anode (NDPA). NDPA shows tensile strength exceeding 1600 MPa, significantly beyond the literature reported values of 60MPa. This strengthening was linked to grain size refinement into the sub-30nm range, in addition to grain texture refinement resulting in only 17% of the slip systems for nickel being active. Anisotropy was observed in nickel deposits, which was linked to texture evolution inside of the coating. Microsample testing guided the selection of 15?m layer of copper deposition followed by a 250 ?m NDPA layer. Classical formulas for structural collapse were used to guide an experimental parametric study to establish a weight/volume efficient strut topology. Length, diameter and thickness were all investigated to determine the optimal column topology. The most optimal topology exists when Eulerian buckling, shell micro buckling and yielding failure modes all exist in a single geometric topology. Three macro-scale sandwich topologies (pyramidal, tetrahedral, and strut-reinforced-tetrahedral (SRT) were investigated with respect to strength-per-unit-weight. The topologies were optimized across length scales using texture on the nano-scale microsamples on the micro-scale, and the parametric column study on the meso-scale. The results showed that additive manufacturing as a viable method for removing geometric constraints observed by other manufacturing methods. The SRT was the most optimized topology showing the highest strength-per-unit-weight. The final topology sits in a best-of-both areas of material space exceeding the commercially available honeycombs strength per relative density by 1670%