32584 research outputs found
Sort by
Embedding-Driven Synthetic Malware Generation with Autoencoders and Cluster-Tangent Diffusion
No full textMalware has become increasingly sophisticated over the years, with zero-day attacks emerging at an alarming pace. Effective detection and analysis demand real malware samples, which are expensive and skill-dependent to extract. As a result, generating high quality synthetic samples from scarce data sets becomes a crucial method for strengthening detection software. This paper focuses on presenting generation techniques that optimize the embedding space to produce high-quality synthetic samples, even under constrained datasets. The dataset used in this paper consists of 500 Windows malware API call samples that were processed using embedding and Generative AI (Gen AI) techniques to generate synthetic malware. Two novel contributions are highlighted in this paper. (1) The integration of autoencoders with pretrained NLP models (BERT and ELMo) to enhance the quality of embeddings. Autoencoders extract features and learn patterns from the data to generate higher-quality embeddings than those generated using other techniques alone. (2) Cluster-Tangent Diffusion (CT-Diff): a novel application of manifold diffusion. Manifold diffusion improves upon diffusion and other Gen AI techniques by focusing on generating samples along the distribution of the original data using structured noise instead of standard gaussian noise. Collectively these two contributions have consistently outperformed previous techniques. Furthermore, the results demonstrate the feasibility of generating reliable fake samples even in low data scenarios
AISStream-MCP: A Real-Time Memory-Augmented Question-Answering System for Maritime Operations
No full textPorts and maritime operations generate massive real-time data streams, particularly from Automatic Identification System (AIS) signals, which are challenging to query effectively using natural language. This study proposes a prototype AISStream-MCP, a memory-augmented real-time maritime question-answering (QA) system that integrates live AIS data streaming with a Model Context Protocol (MCP) toolchain to support port operations decision-making. The system combines a large language model (LLM) with four MCP-enabled modules: persistent dialogue memory, live AIS data query, knowledge graph lookup, and result evaluation. We hypothesize that augmenting an LLM with domain-specific tools significantly improves QA performance compared to systems without memory or live data access. To test this hypothesis, we developed two prototype systems (with and without MCP framework) and evaluated them on 30 queries across three task categories: ETA prediction, anomaly detection, and multi-turn route queries. Experimental results demonstrate that AISStream-MCP achieves 88% answer accuracy (vs. 75% baseline), 85% multi-turn coherence (vs. 60%), and 38.7% faster response times (4.6 s vs. 7.5 s), with user satisfaction scores of 4.6/5 (vs. 3.5/5). The improvements are statistically significant (p \u3c 0.01), confirming that memory augmentation and real-time tool integration effectively enhance maritime QA capabilities. Specifically, AISStream-MCP improved ETA prediction accuracy from 80% to 90%, anomaly detection from 70% to 85%, and multi-turn query accuracy from 65% to 88%. This approach shows significant potential for improving maritime situational awareness and operational efficiency
Equity in Public Budgeting: Community Engagement in Morgan Hill
No full textThrough this research, I will examine what citizen participation strategies have been implemented and their role and effectiveness in addressing wealth inequities. My research will explore these topics and examine what programs and policies the City of Morgan Hill can implement to increase residents\u27 sense of belonging and ultimately push for greater social equity. My primary research question is, what are the programs and policies that the City of Morgan Hill can implement to engage immigrant communities, specifically Spanish-speaking residents, and low-income residents in shaping decisions around public investments and funding? This project will serve as a needs assessment to better understand the actions that Morgan Hill could take to better engage its immigrant and low-income residents, critical communities in the city, in its public budgeting process. Specifically, I will examine the social and political considerations that may act as facilitators or barriers to implementing these programs and policies. In addition, I will use a comparative lens to examine what strategies other regional municipalities, such as Redwood City, Mountain View and San Jose, have put into practice to increase community engagement related to the distribution of public funds. By exploring the strategies and related challenges and opportunities of regional local governments, I will better understand the possibilities in Morgan Hill
Reproduction and Embryonic Development of Monkeyface Prickleback Cebidichthys violaceus in Captivity
No full textUnderstanding the sequence of embryonic and larval development and the factors necessary to induce reproduction in captivity are critical for developing new species for commercial or conservation aquaculture. In this study, we describe the adult reproductive behaviours and development of eggs, embryos and early larvae of captive monkeyface pricklebacks, Cebidicthys violaceus, compared to previously documented wild observations. Eggs were laid in cohesive clutches and guarded by the male parent until hatching began 23 days post fertilization at 13°C. Fertilized eggs were spherical, approximately 1.5 mm in diameter, covered in an opaque chorion, and contained six adhesive pads around the outside. We characterized the rate of depletion of yolk and the oil globule and growth of the embryo from fertilization until hatching. Notable embryonic stages were documented, including the timing of the first heartbeat, and the development of otoliths, intestinal tract, eye pigmentation, mouth, fins and the circulatory system. Larval length at hatching was about 7.4 mm, and larvae were immediately mobile and feeding on live food. Larvae were cultured and observed up to 18 days post hatch
Affective polarization in Latin America
No full textStudies on the determinants and consequences of affective polarization have proliferated in the past decade, focusing mostly on the US and Western European countries. Empirical and anecdotal evidence presented in this chapter, however, suggests that such a phenomenon is also prevalent in many Latin American countries, which may have important consequences for the stability and survival of democracy in a region where partisan institutions tend to be weak. In this chapter, we present a critical review of the scant literature on affective polarization in Latin America. We then carefully analyze levels of partisan affective polarization (PAP) and leader affective polarization (LAP) across the region, as two different yet connected levels of analysis that require more attention in future academic research. Our findings in the region indicate that scholars should devote more attention to understanding the causes and consequences of LAP—as opposed to PAP—in contexts where parties and party systems are weak. We conclude by inviting scholars to think \u27outside the box\u27 to devise strategies that allow us to conceptualize and measure this phenomenon in developing democracies, as well as by discussing several research avenues that scholars can pursue to advance our understanding of this phenomenon in Latin America
Forecasting Pediatric Emergency Department Arrivals: Evaluating the Role of Exogenous Variables Using Deep Learning Models
No full textBackground Forecasting pediatric emergency department (ED) demand remains a critical challenge in healthcare operations. This study aimed to identify exogenous variables influencing pediatric ED visits and evaluate the performance of different forecasting models. Method Using a retrospective observational design, we analyzed 192,347 pediatric ED visits across nine hospitals in Southeast Michigan between 2017 and 2019. Patient data were aggregated into daily arrival counts and enriched with exogenous variables such as weather, air quality, pollen, calendar, Google search trends, and chief complaints. Feature selection was performed using XGBoost and SHapley Additive exPlanations to identify the most influential predictors. Three forecasting models were developed: a Naïve baseline, Long Short-Term Memory (LSTM), and an attention-based neural network. The models were evaluated across 1-day, 7-day, and 14-day forecasting horizons using mean absolute percentage error (MAPE) and R2 metrics. Results LSTM and attention-based model significantly outperformed the Naïve baseline across all horizons. The LSTM model incorporating calendar data achieved the best 1-day forecast (MAPE: 8.71 %, R2: 0.67). For 7-day forecasts, the attention-based model using chief complaint data performed best (MAPE: 9.18 %, R2: 0.57). At 14 days, the attention-based model without exogenous inputs outperformed most LSTM variants, reflecting superior performance in long-range forecasting. Among exogenous variables, calendar and chief complaint data added the most predictive value, while Google Trends and pollen data introduced noise and diminished model performance. Conclusion Combining deep learning architectures with selected external data improves pediatric ED arrival forecasting. From an operational perspective, such forecasts can support more efficient staffing, reduce wait times, and mitigate ED crowding
Integration of UAV and Remote Sensing Data for Early Diagnosis and Severity Mapping of Diseases in Maize Crop Through Deep Learning and Reinforcement Learning
No full textHighlights: What are the main findings? A framework combining UAV, satellite, and weather data to detect crop diseases. Ensemble deep learning and reinforcement learning improve classification and severity mapping. What is the implication of the main finding? Enables near real-time, scalable crop disease management to reduce yield losses. Early detection and hotspot mapping optimize water and pesticide use for sustainability. Accurate and timely prediction of diseases in water-intensive crops is critical for sustainable agriculture and food security. AI-based crop disease management tools are essential for an optimized approach, as they offer significant potential for enhancing yield and sustainability. This study centers on maize, training deep learning models on UAV imagery and satellite remote-sensing data to detect and predict disease. The performance of multiple convolutional neural networks, such as ResNet-50, DenseNet-121, etc., is evaluated by their ability to classify maize diseases such as Northern Leaf Blight, Gray Leaf Spot, Common Rust, and Blight using UAV drone data. Remotely sensed MODIS satellite data was used to generate spatial severity maps over a uniform grid by implementing time-series modeling. Furthermore, reinforcement learning techniques were used to identify hotspots and prioritize the next locations for inspection by analyzing spatial and temporal patterns, identifying critical factors that affect disease progression, and enabling better decision-making. The integrated pipeline automates data ingestion and delivers farm-level condition views without manual uploads. The combination of multiple remotely sensed data sources leads to an efficient and scalable solution for early disease detection
Study on Stage Characteristics and Multi-Factor Optimization Regulation of Performance of Ice Thawing Agent in Low Temperature Environment
No full textDe-icing agents play a crucial role in winter road maintenance, yet their excessive application can result in pavement deterioration and environmental issues. Existing dosage guidelines lack comprehensive data on the dynamic response of de-icing agents under low-temperature conditions, particularly regarding stage-specific characteristics and multi-factor interactions. This research systematically evaluated the effectiveness of four de-icing agents (NaCl, CaCl2, MgCl2, CH3COOK) within a temperature range of −5 °C to −25 °C, elucidating the two-phase ice-melting process (solid-phase followed by salt solution de-icing) with distinct kinetic mechanisms—a previously underexplored temporal pattern. The study quantified the differential impacts of particle size (small-particle CaCl2 exhibiting 12% higher efficiency than sheet-like forms), dosage linear correlation, and negligible effects of ice layer thickness and road surface composition, which have not been systematically validated in prior studies. Temperature sensitivity was further refined: NaCl showed a 42.4% efficiency drop between −5 °C and −25 °C, while MgCl2 maintained stable performance, supporting its potential as an environmentally sustainable alternative. This work provides a quantitative basis for dynamic dosage regulation by integrating stage characteristics and multi-factor optimization, addressing gaps in existing guidelines
On the Road to a Better Life? Rural Road and Economic Development in Albania
No full textThis study evaluates the impact of investment in rural roads on household welfare in Albania using a difference-in-differences method. We find that treated households experienced a 35-percentage point increase in the road quality and reduced travel times to the nearest motorable roads. Land and housing values rose significantly, and household heads were less likely to be unemployed, with a shift toward self-employment. While household income did not change significantly, consumption—particularly on non-food items—increased. These findings highlight the welfare-enhancing effects of improved connectivity and suggest that rural infrastructure investments can promote local economic activity. Policymakers should consider pairing road upgrades with complementary interventions, such as access to credit or small enterprise support, to maximize long-term gains
