22505 research outputs found
Sort by
The Impact of Screen Time Patterns on Digital Behaviour and Productivity: A Data Analytics Study Across Age Groups
No full textThe increasing incidence of screen time of all ages has generated a greater interest in the effect of screens on the human psyche. This paper looks at behavioural data of 2, 000 individuals aged 13-64 to investigate the predictability of a composite mental- health score by various modalities of screen-time, lifestyle and psychological charac- teristics. We determine, based on exploratory statistics, multivariate regression and su- pervised machine-learned models, that in as much as the influence of total screen time and social-media consumption has significant relationships with worse mental health, the effects are small when juxtaposed to those of lifestyle and emotional variables. The duration of sleep and physical activity can be called a robust protective predictor, whereas the strongest indicators of lower mental-health scores are stress, anxiety, and symptoms of depression. The regularized linear models are better than tree-based and distance-based models (maximum performance: RMSE = 25.85; R2 = 0.34). All in all, the findings indicate that psychological and lifestyle variables tend to influence the development of the mental-health outcomes more significantly than the use of a screen. Not all screen-time restrictions might be effective but some specific strategies, like enhancing sleep hygiene and encouraging balanced digital habits, could be even more effective. The next generation of studies needs to combine longitudinal information and objective data of digital traces and more accurately reflect the changing trends of digital interaction
On the Use of Nonlinear Electrophoresis for Separating Particles and Cells in Insulator-Based Electrokinetic Systems
Get PDFThe increasing demand for efficient separation processes of micron-sized particles, including microorganisms, highlights the importance of developing low-cost, non-labor-intensive, and time-efficient techniques. Insulator-based electrokinetic (iEK) systems have proven to be a practical alternative to bench-scale methodologies such as centrifugation and membrane filtration. Several studies have reported the separation of microparticles by employing direct current (DC) voltages. However, all of them have overlooked the phenomenon of nonlinear electrophoresis (EPNL) and its role in the electrokinetic migration of particles. This study combines mathematical modeling and experiments to extend the capabilities of iEK systems. This dissertation has four aims: (1) to separate microparticles and cells using iEK systems, (2) to improve iEK device designs for charge- and size-based separation, and (3) to employ EPNL for reversing migration order of analytes in EK separations, and (4) to develop an empirical equation for particle retention time based on particle characteristics, microdevice features, and applied DC voltages, with a focus on EPNL. In the first aim, electrokinetic separations of mixtures of micron-sized particles, bacteria, and yeast cells was conducted, achieving the challenging separation of highly similar microparticles, a significant contribution to the field. By combining modeling and experimental approaches, two distinct types of microparticles (~5 µm) with a small zeta potential difference of 3.6 mV were successfully separated. The promising findings from this study paved the way for separating microorganisms even in the same domain and genus. Also, this aim presents cell separation in iEK systems where three different binary mixtures of cells (Escherichia coli vs. Saccharomyces cerevisiae, Bacillus cereus vs. Saccharomyces cerevisiae, Bacillus cereus vs. Bacillus subtilis) were separated, with the difficulty level increasing in each subsequent separation. The second aim is to improve the device design for separating a more complex sample, a tertiary sample of microparticles. By combining mathematical simulations and iEK separation experiments, iEK systems were improved for more efficient charge-based and size-based separation. In this study, two configurations were selected for the effective separation of complex samples. The third aim is to reverse the migration order of analytes by altering the electrokinetic regime in electrokinetic microsystems by changing the applied DC voltage. In this study, a charge-based separation of particles was performed in the linear electrokinetic regime, and within the same microdevice, a size-based separation was performed under the nonlinear electrokinetic regime, marking the first reported reversal in elution order in iEK systems. Also, in a different study, the improved design found in Aim 2 was tested to reverse the elution order cells of the same domain and genus but different strains (two strains of Saccharomyces cerevisiae). The successful reversal of elution order in two strains of Saccharomyces cerevisiae highlighted the efficiency of the improved design. Aim four focuses on gathering the findings obtained in Aims 1 to 3 and analyzing the information to develop correlations based on the electrokinetic characteristics of particles and device configuration to predict parameters such as retention time in the separation of (bio)particles. This aim introduces a method to predict particle retention time in iEK systems under linear EK conditions while considering EPNL, particle properties, electric fields, and microdevice features. Experiments with eight distinct microparticles (3.6–11.7 µm, ~-20 mV, and ~-30 mV zeta potentials) were conducted in three iEK microdevices: a device with asymmetrical insulating posts, one with symmetrical insulating posts, and one without posts (postless). The microparticle retention times tR,e were measured at 400–1450 V, leading to three empirical equations for particle velocity, incorporating linear and nonlinear effects. Validation with control particles showed a fair prediction error, highlighting the equations\u27 potential for designing particle separations in iEK systems. This dissertation highlights the capability of iEK systems to achieve highly discriminatory separations of mixtures containing micron-sized entities, including microparticles and cells
Crafting Human-AI Collaborative Analysis for Usability Evaluations
Get PDFIn the rapidly evolving domain of human-computer interaction (HCI), ensuring a high-quality user experience (UX), which encompasses all aspects of a user’s interaction with an interface, is critical to the success and adoption of technology. Central to this is the concept of usability: the degree to which an interface can be effectively, efficiently, and satisfactorily used by its intended audience. Usability testing is a key method for uncovering usability problems, relying on structured evaluations involving real users. However, traditional approaches to analyzing usability test recordings—where UX evaluators observe user behavior and verbalizations while taking notes—are labor-intensive and prone to human error. These challenges are compounded by the well-documented evaluator effect, whereby individual differences among evaluators lead to inconsistent identification and interpretation of usability problems. This dissertation addresses these limitations by proposing a nuanced human-AI collaborative approach to enhance the effectiveness of usability analysis. To identify optimal configurations of such collaboration, I examined four key factors: 1) Representations of AI, comparing non-interactive visualizations, user-directed (passive) conversational assistants (CAs), and system-directed (proactive) CAs; 2) Interaction modalities, contrasting voice- and text-based interactions when engaging with CAs; 3) Timing of suggestions, analyzing the impact of AI-generated suggestions delivered before, during, or after the occurrence of a usability problem; and 4) Perceived expertise, investigating differences in collaboration with CAs simulating novice versus expert UX evaluators. The fourth study also explored how UX evaluators\u27 behaviors and attitudes evolve through long-term collaboration with a CA. This body of work culminates in a comparative evaluation of the usability results produced by human-only, AI-only, and human-AI collaborative analysis. The results demonstrate that well-designed human-AI collaborative configurations yielded significant improvements in both the quality of usability results and the analysis experience. The contributions of this dissertation include: empirical insights into the practices and challenges of usability analysis; evidence on how interaction modality, suggestion timing, and perceived expertise affect evaluators’ analytical behavior and perceptions; AI-powered analysis tools for identifying usability problems; a dataset capturing the types of questions UX evaluators ask of CAs; and a methodology for evaluating the quality of usability results. Collectively, these contributions support a framework for nuanced human-AI collaboration for usability evaluations. The thesis of this dissertation is: Nuanced human-AI collaboration improves the effectiveness of usability evaluations compared to traditional non-AI methods when drawn on evidence from user testing on representations of AI, interaction modalities, timing of suggestions, and perceived expertise
From Contact to Care: A Comprehensive Analysis of Mental Health Practices Across the Criminal Justice Continuum
Get PDFThis capstone attempts to gain a further understanding of the state of mental health training, protocols, and practices across the continuum of the criminal justice system. The various sections of this paper will look at law enforcement, corrections, and court/community and their roles in mental crisis training and interaction. As the topic of mental health is something increasingly discussed in the present day, it is also something that is in need of more research, in order to create higher standards when it comes to individuals with mental health concerns and ensuring that they receive the proper levels of care that they deserve. This paper attempts to contribute to the gaps in research on this particular topic, and to create a comprehensive look at mental health in the criminal justice system as a whole, analyzing existing policies and techniques, while shedding light on many aspects to be improved upon as well
Exploring Kosovo\u27s Need for a Technology-Driven Offset Strategy
Get PDFAt a time where technological innovation is redefining national security, Kosovo risks falling behind by failing to adopt emerging technologies for defense and security. This research addresses that vulnerability by exploring the need for a dedicated national strategy as a solution. Drawing on global best practices, including the United States Offset Strategy, NATO frameworks, and innovation models from the Baltic countries, the study outlines a potential path forward. Additionally, through qualitative analysis, including interviews with key stakeholders across public, private, and academic sectors, this study looks into the opportunities and challenges of developing a context-specific strategy for integrating emerging technologies into Kosovo’s security architecture. Central to this proposed strategic framework are public-private partnerships, legislative reforms to foster innovation, and the establishment of a Defense National Innovation Agency to coordinate and drive defense innovation. With the right direction, Kosovo can turn its tech-savvy population into a national asset by strengthening national security, improving defense readiness, and shaping its role as a credible regional actor in defense innovation
Parsing of Math Formulas and Chemical Diagrams using Graph-Based Representation and Attention Models
Get PDFMathematical formulas and chemical diagrams appear frequently in scientific documents but are often embedded as visual content, either rasterized or vector-based images, limiting their accessibility and automated analysis. This thesis aims to bridge this gap by presenting a graph-based visual parsing framework that recognizes and parses these notations from both vector and raster image inputs in digital documents. For mathematical formulas in born-digital PDFs, we construct Symbol Layout Trees (SLTs) using a graph defined over vector-based primitives, capturing spatial relationships, avoiding relying on OCR. For born-digital chemical diagrams, we introduce a Minimum Spanning Tree (MST)-based technique that extracts molecular structure graphs by interpreting vector graphics using domain-specific spatial and symbolic constraints. To parse rasterized images, we develop a multi-task, segmentation-aware neural network that operates on over-segmented visual primitives extracted via line segment detection and watershed-based segmentation. We create annotated training data by aligning vector-based ground truth with detected visual primitives in raster images. The model jointly performs symbol classification, segmentation, and relationship classification in a multi-task learning framework, utilizing discrete attention mechanisms to dynamically modify input features over iterative passes. We enhance robustness using synthetic structural and visual noise applied at the primitive level to simulate degradations in real document images and mitigate class imbalance through stratified sampling and loss reweighting strategies, including weighted cross-entropy, class-balanced and focal losses. We introduce a two-stage graph attention model to support cross-task learning, where class distributions from the first stage are used to inform refinement in the second. Evaluation metrics compare nodes and edges in the predicted graphs to ground truth using adjacency matrices and Hamming distances to quantify structural and labeling errors. The results and analysis across mathematical and chemical datasets show that (1) input line-of-sight (LOS) graph representation improves expression coverage (the upper bound on the number of expressions that can be correctly parsed) and reduce number of edge hypotheses for math, while 6 nearest-neighbor (6NN) graphs are better suited for chemistry due to their local structure, (2) attention mechanisms and cross task interaction enhance structural prediction, (3) primitive-level noise augmentation and loss rebalancing and aggregation improve generalization across input conditions. Together, these findings support the development of a unified and extensible framework for visual parsing of structured scientific notations across domains
Comprehensive Investigation of the Torsional behavior of Triply Periodic Minimal Surface (TPMS) inspired Structures
Get PDFThis work is concentrated on research of torsion properties of TPMS structures, numerically and experimentally. Currently, there are not many works that concentrate on studying torsion properties of TPMS structures. Most of them are concentrated on structures for biomedical applications and structures, printed of Ti6Al4V titanium alloy. In other areas, such as aerospace, automotive, etc. TPMS structures may also find applications. In this work, shear modulus, shear strength and resilience are tested, depending of the following conditions: cell geometries Diamond, Gyroid and Primitive, with relative densities of 30%, 50% and 70%, and unit cell sizes of 10 mm, 15 mm and 20 mm. Sample has a diameter of 20 mm and length of 40 mm. Size of all the samples is the same. For the study, full-factorial design is used, meaning that there all 27 possible conditions are tested. Simulation is done in Abaqus 2022 software. Mesh convergence study is done to define optimal mesh size. Material properties obtained, based on paper review. Test is displacement driven, meaning that torque is applied to the sample and from that, stresses are calculated. Reference points are added to represent the surfaces. One edge was fixed in all translations and rotations. Other was fixed in translations, perpendicular to the sample axis, but axial translation was allowed to prevent the appearance of additional tensile loads on the sample. Rotations in plane with the sample axis were also restricted. Twist is applied around the sample axis. Then, simulation is ran. After that, results are viewed and torque-revolution curves are obtained. To obtain stress-strain curves, minimum moment of inertia is measured on the models. After that, stress-strain curves are calculated. Regarding Von Mises stress distribution, Diamond and Gyroid had an even distribution, while Primitive had areas, where stress was concentrated. Smaller number of unit cell size (more unit cells in the sample) leads to a more even stress distribution in the sample. Regarding effect of cell geometry, Dimond and Gyroid have shown almost identical stress-strain curves, in different configurations one of the samples had slightly higher shear yield strength and shear modulus. Primitive had the highest shear modulus and shear yield strength, with unit cell size of 10 mm and 15 mm. Regarding the effect of relative density, for most samples with its increase, shear modulus and shear yield strength increased. The exception was Primitive, where shear modulus and shear yield strength decreased with the increase of relative density. Regarding the effect of unit cell size, Diamond and Gyroid samples have shown mostly identical stress-strain curves, while Primitive has shown significant difference in properties with change in unit cell size. With the decrease of unit cell size from 20 mm to 15 mm, shear modulus and shear yield strength increased, while further decreased for 10 mm unit cell size. This difference in values increased with decrease of relative density. Regarding resilience, Primitive samples had higher values on average, than Gyroid and Diamond, who had slightly differentiating properties for different configurations. Primitive sample with relative density of 30% and unit cell size of 15 mm has shown the highest resilience, significantly higher than other samples. In summary, Diamond and Gyroid have shown less differentiating properties than Primitive. Also, stresses in Primitive were not as evenly distributed, as in Gyroid and Diamond. For Primitive, increase in relative density and decrease in unit cell size caused more even stress distribution. When there are two or more interconnections between unit cells in redial direction, stress is more evenly distributed. Gyroid and Diamond have shown properties, suitable for torsion application. Primitive can also show good properties in torsion application, but only when there are two or more unit cells in radial direction
Product, Process, and Behavior Influences on Value Retention Processes: Toward a Circular Economy in Consumer Electronics
Get PDFConsumer electronic products (CEP) manufacturing and retail contribute markedly to global industrial activity; worldwide estimates suggest revenue growth from 1.5 trillion by 2026. Consequently, the production, distribution, use, and end-of-life (EOL) disposition of these products are responsible for considerable social and environmental impacts, including 35 million metric tonnes of waste to landfill per year, 793 million metric tonnes (MMT)—and growing—of carbon dioxide-equivalent (CO2e) greenhouse gas (GHG) emissions per year, and innumerable adverse effects on human health and development. Driving these impacts are the inherent characteristics of CEPs themselves, which increasingly require critical materials in design, high energy consumption in manufacturing, and complex treatment of hazardous waste streams at EOL. In the age of information, however, access to CEPs is increasingly essential to technological and socioeconomic development, particularly in emerging Global South economies. Serving these growing consumer needs while mitigating environmental and human health impacts is a considerable challenge. The circular economy is broadly emerging as a means to address these challenges. In particular, value retention processes (VRPs) including remanufacturing, refurbishing, repair, and direct reuse are gaining market share and acceptance as practical applications of circular economy principles. In some industry sectors—e.g., automotive, aerospace, and commercial machinery—VRPs have been found to be more economically efficient than and environmentally preferable to incumbent linear business models, offering a means to decouple economic advancement from increasing environmental impact. Accelerating demand for CEPs and accordingly growing materials, energy, manufacturing byproducts, and EOL waste problems highlight the necessity of such decoupling to sustainable development in this sector as well. To that end, this research provides a framework for quantifying the market potential for and possible impacts of a shift to VRPs in consumer products industries at large, using the CEP sector as a high-impact case study. To this end, Chapter 2 conducts product-level material flow analysis (MFA) for five key CEP types to assess the material and behavioral feasibility of VRP models in the CEP sector across developmental strata, highlighting the United States of America (US) and the Republic of Ghana (GH) as Global North and South case studies, respectively. Chapter 3 then assesses the relative environmental performance of possible VRP business models at the product level, using survey data on specific rates and modes of failures to inform new VRP life cycle assessment (LCA) models for two case study products at opposite ends of the CEP spectrum: a smartphone and a domestic refrigerator. Finally, Chapter 4 assess the economic viability of VRP business models in the CEP sector through the user lens, proposing a new model for multigenerational multicriteria decision analysis (MCDA) between new and VRP CEPs. We evaluate this model using best available market, survey, and product specification data, and investigate how product attributes and consumer preferences influence VRP purchasing decisions in CEP markets. Outcomes of this research are twofold. First, Chapter 2 and 3 case study modeling results suggest that VRPs are indeed materially feasible and environmentally preferable across the CEP sector, even under current market, technology, and behavioral conditions. Similarly, Chapter 4 results illustrate that circular shifts are a plausible economic reality, and sensitivity analyses highlight strategies to optimize the benefits of and alleviate barriers to such shifts. Second, the underlying modeling frameworks themselves provide a foundation for quantitative analysis of how technology, market, and policy factors affect the viability and preferability of VRPs across contexts. This architecture is thus adaptable across product types, market circumstances, and developmental strata, supporting broader global analysis of circular opportunity, enabling conditions, and possible benefits
Advancing the Utility of Unmanned Aerial Systems (UAS)-Based Imaging Techniques in Broadacre Agriculture: A Multimodal Case Study on Table Beets
No full textEfficient and sustainable food production and management are growing concerns in the context of an ever-increasing global population. It is in this context that the integration of remote sensing with advanced imaging technologies presents transformative opportunities for data-driven decision-making in agriculture. This study therefore explores the use of unmanned aerial systems (UAS) equipped with multispectral, hyperspectral, and LiDAR sensors for the non-destructive monitoring of table beet (Beta vulgaris) crop traits, with a specific focus on root yield estimation and foliar disease assessment. Table beet, a subterranean crop of increasing commercial and nutritional importance, poses unique challenges for above-canopy sensing due to its below-ground storage organ. Two seasons of UAS campaigns were conducted (2021 and 2022) at Cornell AgriTech in Geneva, NY, capturing multispectral (475, 560, 668, 717, and 840 nm), hyperspectral (400-1000 nm), and LiDAR data across multiple growth stages. Initial analysis employed hyperspectral imagery to identify key narrow-band wavelengths predictive of root yield, yielding leave-one-out cross-validation R2 values between 0.85–0.90 and RMSE values of 10.81–12.93%. The 760–920 nm spectral range was most indicative of yield performance. Subsequent modeling efforts focused on developing growth stage- and season-independent yield prediction models. A Gaussian Process Regression model, using only multispectral data, achieved an R2test = 0.81 and MAPEtest = 15.7%, while the fusion of hyperspectral and LiDAR data produced an R2test = 0.79 and MAPEtest = 17.4%. This investigation additionally revealed the added value of structural information. For disease monitoring, we developed various machine learning models using features derived from five-band multispectral imagery, including vegetation indices and gray-level co-occurrence matrix (GLCM)-based texture metrics. The top-performing model achieved R2test = 0.90 and RMSEtest = 7.18%, and hyperspectral-based models reached R2test = 0.87 and RMSEtest = 10.1%. Here we demonstrated disease severity monitoring at relatively course resolution. This work demonstrates the efficacy of UAS-based multimodal sensing for high-throughput phenotyping, while also addressing key questions in sensor selection, feature engineering, and model generalization. The methodologies developed here offer scalable, data-driven solutions for yield forecasting and disease assessment, and may be adapted for broader use across other crops and sensing platforms
Amorphous Oxide Semiconductor Materials and Devices for Monolithic and Heterogeneous Integration
Get PDFThe expanding consumer electronics industry has spurred significant advancements in display technology. Thin-Film Transistors (TFTs) serve as active matrix switching components in LCD and OLED panels. Amorphous metal-oxide semiconductors (AOS) support large-area deposition at low temperatures and boasts an electron mobility many times greater than that of amorphous silicon. This work presents a comprehensive study on AOS materials and devices for their introduction to display, monolithic integration, and heterogeneous integration applications. Key studies have focused on advancing the state of (Indium Gallium Zinc Oxide) IGZO TFTs by addressing challenges in device uniformity, reliability, and modeling. Device uniformity was improved by modifying process parameters to allow for higher degree of film uniformity during deposition. Devices fabricated with this modified process demonstrate exceptional resistance to the application of traditional bias stress. Application of intensive bias and illumination-bias stress treatments led to distinctive transfer characteristics, differing in shift magnitude, distortion, and hysteresis behavior. Silvaco TCAD and a previously defined mobility model were used to simulate this behavior and explore the defect states created during intensive bias stress. Utilizing ion implantation for self-aligned source/drain regions present a path towards sub-micron device scaling. Past reports have demonstrated boron implanted self-aligned TFTs with excellent on-state and off-state performance. However, when subjected to thermal stresses above 175ºC the device transfer characteristics gradually shift over time. From this work it is hypothesized that this instability is related to the implanted boron dose, with higher doses presenting more shifting. Interpretation of electrical results suggests that boron can exist in two states: an active form that bonds with interstitial oxygen, increasing oxygen vacancies and enhancing conductivity, and an inactive form as an isolated interstitial atom. At low boron doses, the active state is dominant, improving conductivity, whereas at higher doses, the inactive state prevails, leading to reduced current and, in extreme cases, charge injection degradation. A thermally-activated diffusive mobility and percolation theory are two contending processes that have been proposed to govern electron transport in IGZO. This work builds upon previous investigations on the temperature dependence of channel mobility in IGZO TFTs, where transport behavior from 170 K to room temperature (RT) is clearly described by a thermally-activated diffusive mobility. The isolation of thermally dependent mechanisms via TCAD enabled the separation of the intrinsic and extrinsic components of the observed field-effect mobility. The methods used resulted in a quantitative assessment of the thermally-activated diffusive mobility and the free/total charge ratio. These advancements allowed for the development of a platform to realize the heterogeneous integration of µLEDs on an IGZO TFT backplane. µLEDs were successfully transferred onto an IGZO TFT backplane using a micro transfer printing technique employing a PDMS stamp. Metal deposition techniques were investigated for post passivation anneal interconnects between printed µLEDs and the completed IGZO backplane. Emission of single pixel and RGB pixel circuits was confirmed and a new RGB pixel was designed to minimize pixel cross-talk and sub-pixel leakage. While IGZO is the only AOS technology that has matured enough for commercialization, the electron channel mobility (≈10 cm2/Vs) presents a limitation in its application in back-end-of-line (BEOL) monolithic and heterogeneous integration. As such, alternate candidate AOS materials exist that exhibit channel mobilities 2-3x higher than that of IGZO. Studies in alternate AOS materials such as Indium Tungsten Oxide (IWO), Indium Tin Gallium Oxide (ITGO), and Indium Gallium Zinc Tin Oxide (IGZTO) have been conducted. The investigation on IWO TFTs revealed an unusual metastable device behavior dependent upon annealing temperature. This is believed to be the first report of such behavior, as published works adhere to either a low-temperature or high-temperature regime. The investigations into ITGO and IGZTO serve as preliminary studies; device characteristics support the claims of high channel mobility; however, the influence of defect states clearly indicates the need for further process development. The advancements realized in IGZO TFTs in this work will serve as a foundation for these alternative AOS materials