32584 research outputs found
Sort by
Vertiport Location Selection and Optimization for Urban Air Mobility in Complex Urban Scenes
No full textVertiports, as dedicated facilities for electric vertical takeoff and landing (eVTOL) aircraft, are essential to ensure the efficiency and sustainability of Urban Air Mobility (UAM). However, UAM infrastructure site selection has become increasingly complex due to limited land availability, complex spatial conditions, and the need to balance multiple objectives. Focusing on passenger-carrying UAM operations, this study proposes a systematic framework for vertiport site selection. First, key factors are classified into high, medium, and low levels across the safety, economic, and social dimensions, forming a modular evaluation system. A GIS-based spatial screening process is developed to identify potential vertiport locations. Subsequently, a variable representing the level of demand satisfaction is incorporated into a progressive coverage model specifically designed for vertiport site optimization. A hybrid algorithm is designed to solve the model. Using Shenzhen as a case study, the proposed approach is validated through real-world data. The results show that vertiport size and spatial requirements significantly influence the selection of suitable land types. High economic constraints may cause facility over-concentration, while setting standards aligned with regional functions better supports equitable access. Locating vertiports in high-demand areas enhances demand satisfaction levels, and both service capacity and range strongly influence overall system performance. These findings provide practical insights for future vertiport planning, promoting the efficient use of urban resources and supporting the successful implementation and sustainability of UAM
Do Sustainability Reports Contain Financially Material Information?
No full textRecent years have witnessed significant growth in corporate sustainability reporting. Yet existing research provides mixed evidence on the information content of these reports for investors. We examine the stock market reaction to the announcement of a sample of US corporate sustainability reports incorporating Sustainability Accounting Standards Board metrics that are intended to provide financially material information to investors. Using standard measures of information content, we cannot find compelling evidence that these reports provide a significant amount of new information to investors. Further analysis of a subset of common metrics indicates that they are either financially immaterial or preempted by traditional financial disclosures. Finally, we show that most firms target their sustainability reports at a broad set of sustainability-oriented stakeholders rather than a narrow set of financially oriented investors
Secondhand Social Capital and Idea Quality in Open Innovation Communities
No full textOpen innovation communities provide valuable opportunities for creators from diverse backgrounds to collaboratively generate and refine new ideas that companies can implement in new products or services. However, many submitted ideas in these communities are underdeveloped or misaligned with companies\u27 expectations, raising questions about what drives high-quality ideas and, more specifically, innovation potential. We focus on the role of secondhand social capital, i.e., the indirect network benefits an idea accrues through feedback from providers who are themselves actively engaged with other influential ideas in the community. While prior research has explored ego-centric or first-hand networks, we extend this work by examining how an idea\u27s position within feedback networks shapes its elaboration and innovation potential. We argue that feedback from highly connected feedback providers confers greater visibility, legitimacy, and alignment with community expectations, thereby enhancing the quality of an idea. Using data from an open innovation platform for vehicle design, we find that secondhand social capital significantly predicts higher-quality ideas. By contrast, traditional measures of network constraint (e.g., closure, structural holes) are not consistently and significantly associated with idea quality. Further, our analysis suggests that when feedback is constructive and encouraging, the effect of secondhand social capital is stronger. Our findings contribute to theory by identifying secondhand social capital as a key mechanism linking network structure and idea quality. More broadly, this research bridges micro-level creativity and macro-level innovation literatures by emphasizing feedback through secondhand social capital as a linchpin connecting idea generation and implementation in decentralized, collaborative environments
UAV-Based Transport Management for Smart Cities Using Machine Learning
No full textHighlights: What are the main findings? A smart transport management system based on UAV data integrating advanced machine learning and deep learning techniques is proposed to enhance road anomaly detection and severity classification. The system employs a comprehensive multi-stage framework, integrating a high-precision obstacle detection model, six specialized severity classification models, and an aggregation model to deliver accurate anomaly assessment, enabling strategic, data-driven road maintenance and enhanced transportation safety. What is the implication of the main finding? A scalable and efficient solution is proposed to enhance road safety and optimize transportation management through intelligent anomaly detection and severity assessment. This framework sets a benchmark for future smart city initiatives by leveraging advanced machine learning techniques for proactive infrastructure maintenance and decision-making. Efficient transportation management is essential for the sustainability and safety of modern urban infrastructure. Traditional road inspection and transport management methods are often labor-intensive, time-consuming, and prone to inaccuracies, limiting their effectiveness. This study presents a UAV-based transport management system that leverages machine learning techniques to enhance road anomaly detection and severity assessment. The proposed approach employs a structured three-tier model architecture: A unified obstacle detection model identifies six critical road hazards—road cracks, potholes, animals, illegal dumping, construction sites, and accidents. In the second stage, six dedicated severity classification models assess the impact of each detected hazard by categorizing its severity as low, medium, or high. Finally, an aggregation model integrates the results to provide comprehensive insights for transportation authorities. The systematic approach seamlessly integrates real-time data into an interactive dashboard, facilitating data-driven decision-making for proactive maintenance, improved road safety, and optimized resource allocation. By combining accuracy, scalability, and computational efficiency, this approach offers a robust and scalable solution for smart city infrastructure management and transportation planning
Uneven Sustainability in Academic Publishing: A Call for Transparency and Collaboration
No full textObjective – To assess the sustainability practices of academic publishers and their alignment with library collection management, and to determine the nature of sustainable partnership opportunities at the intersection of library and publisher practices. Design – A cross-sectional, unobtrusive analysis of existing content. Setting – The academic publishing industry. Subjects – Sustainability practices from 16 international academic publishers selected based on their prominence in Western Libraries\u27 print book acquisitions. Methods – The Green Audit Template, a rubric based on the United Nations’ (UN) Sustainable Development Goals (SDGs) and industry standards and inspired by lifecycle assessment (LCA) approach, was employed to assess various aspects of sustainability practices of the 16 publishers. The assessment relied exclusively on the publishers’ websites and analyzed publicly available reports, corporate policies, and sustainability statements from the websites. The analysis focused on the trends found among all publishers rather than identifying major differences between publishers. Main Results – Notable variability was found in the sustainability practices of academic publishers, with transparency and commitment largely influenced by regional regulations, funding levels, and institutional capacity. Publishers based in the United Kingdom (U.K.) and the European Union (E.U.), where environmental, social, and governance (ESG) reporting is mandatory, demonstrated the highest levels of disclosure, with all of them publicly sharing sustainability initiatives. In contrast, only 66% of North American publishers voluntarily disclosed such efforts, while 31% of all publishers lacked any publicly available sustainability information. Although 44% of publishers provided annual reports detailing environmental commitments, only half of them showed clear progress toward achieving their goals. In terms of sustainability commitments, 38% of publishers pledged to reach net-zero emissions, with target years ranging from 2040 to 2050. However, only two publishers had achieved ISO 14001 certification, an internationally recognized environmental management standard. Additionally, while 63% of publishers had signed the UN SDG Publishers Compact, the extent of their sustainability efforts varied, with some publishers focusing more on advocacy rather than implementing environmentally friendly practices. Material sourcing and production practices also revealed gaps in sustainability efforts. More than half of the publishers, 54%, reported using sustainably sourced paper, and 25% discussed their recycling processes, how they dispose of e-waste, or how they manage edition changes. Very few provided details on Open Educational Resources, servers, sustainable ink and glue. Transportation and infrastructure were also key areas of concern. While 43.75% of publishers reported implementing sustainable travel policies such as reducing business travel and promoting public transit, only 12.5% of publishers disclosed warehouse locations, and just 31.25% addressed ecofriendly shipping practices, primarily through strategies like reducing print production and minimizing plastic packaging. Among the 43.75% of the publishers that acknowledged sustainability efforts in their offices, a variety of practices were reported, such as using renewable energy, reducing in-office printing, using carbon credits, and funding reforestation projects to achieve carbon neutrality. Conclusion – While some academic publishers are making progress toward sustainability, some of their practices vary widely. Government-mandated ESG reporting plays a crucial role in driving disclosure, with U.K. and E.U. publishers showing more comprehensive sustainability commitments compared to their North American counterparts. Larger publishers generally lead in green initiatives, while smaller publishers face challenges due to limited resources and capacity. The need for increased transparency and accountability in publishing is highlighted, and libraries and publishers are encouraged to adopt and adapt the Green Audit Template as a common reporting framework. Further research may delve into digital infrastructure sustainability, recycling practices, the role of smaller publishers in green initiatives, and collaborative efforts between libraries and publishers
Multimodal Emotion Detection and Analysis from Conversational Data
No full text—Emotion recognition in conversations has become increasingly relevant due to its potential applications across various fields such as customer service, social media, and mental health. In this work, we explore multimodal emotion detection using both textual and audio data. Our models leverage deep learning architectures, including Transformer-based models such as Bidirectional Encoder Representations from Transformers (BERT), Robustly optimized BERT approach (RoBERTa), Audio Spectrogram Transformer (AST), Wav2Vec2), Bidirectional Long Short-Term Memory (BiL-STM), and four fusion strategies that combine features from multiple modalities. We evaluate our approaches using two widely used emotion datasets, IEMOCAP and EMOV. Experimental results show that fusion models consistently outperform single-modality models, with Late Fusion achieving the highest weighted F1-Score of approximately 78% on IEMOCAP using both audio and text
From Silos to Synergies: How to Enhance Interdisciplinary Collaboration Among Graduate Students
No full textInterprofessional education (IPE) is essential for preparing collaborative healthcare practitioners, yet opportunities at San José State University (SJSU) have been limited, inconsistent, and largely dependent on individual faculty initiative. This capstone project aimed to strengthen IPE at SJSU by assessing institutional capacity and developing feasible, sustainable strategies to enhance exposure and participation across health-related programs. Guided by the Person–Environment–Occupation–Performance (PEOP) model and adult learning theory, the project combined a needs assessment, literature review, and partnership with the Healthy Development Community Clinic (HDCC) to create an instructor implementation guide and student introductory modules for HDCC service-learning onboarding. An IRB-approved mixed-methods faculty survey used the Interprofessional Education Collaborative’s Institutional Assessment Instrument and open-ended questions to examine the institution’s capacity for IPE. Additionally, a pilot student workshop was implemented for applied program development to explore engagement, role clarity, and perceived value of IPE. The findings from both the workshop and survey were used to refine the guide and modules. Survey results indicated low institutional capacity and limited framework integration despite strong faculty recognition of IPE’s benefits, while workshop observations and feedback demonstrated intrinsic student interest in collaborative learning, high engagement, increased understanding of roles, and a desire for additional IPE experiences. Integrating these findings into the guide and modules resulted in scalable, contextually relevant capstone deliverables. These deliverables were designed to enhance institutional coordination, support faculty efforts, and prepare students for team-based, person-centered care, while positioning occupational therapy as a leader in advancing IPE at SJSU
Diffusion Model On API Call Classification
No full textMalware classification through Application Programming Interface (API) call analysis is essential for modern cybersecurity. However, traditional classification approaches often have to face significant challenges due to limited and imbalanced datasets. Therefore, this project proposes a class-conditional diffusion model designed to generate realistic synthetic API-call embeddings that can be trained based on classes and generate realistic malware API call embeddings for data augmentation. Furthermore, seven embedding techniques are explored: Bag of Words (BoW), TF- IDF, Word2Vec (Skip-gram and CBOW), FastText, Doc2Vec, and DistilBERT. The two best synthetic embeddings will then be compared with the corresponding embeddings generated from Wasserstein GAN with Gradient Penalty (WGAN-GP), another popular generative model. These synthetic embeddings are evaluated through downstream classification performance using Gaussian Naive Bayes, Random Forest, Support Vector Machine (SVM), and Multi-Layer Perceptron (MLP) at two levels, which are 7 malware families and 11 malware categories. Final results demonstrate that mixing synthetic data generation improves classification accuracy by up to to 8.1%. WGAN-GP outperformed diffusion for high-dimensional BoW embeddings, while diffusion showed advantages for low-dimensional TF-IDF embeddings. Optimal ratios ranged from 90% original -10% synthetic to 60%-40% depending on embedding type. BoW and TF-IDF embeddings showed the most consistent improvements. These findings demonstrate that generative model selection should be guided by embedding and dataset characteristics in data-limited scenarios
LAYER-SPECIFIC PERTURBATIONS FOR GENERATING MORPHENCE STUDENTS IN TIME-SERIES MOVING TARGET DEFENSE
No full textTime series forecasting models are vulnerable to adversarial perturbations. Even the smallest input modifications can produce significantly erroneous forecasts. Moving Target Defense (MTD) methods address this vulnerability by introducing controlled model diversity at inference time. In this work, the Morphence framework is extended to regression based forecasting to evaluate how different student model perturbation strategies can influence adversarial robustness. A Transformer model is used as the base, and then multiple student models are created through structured parameter perturbations. Two unique ensembles of students are then examined. The first is a vanilla Morphence style ensemble produced through small stochastic weight changes. The second is a novel ensemble generated via stronger and more diverse perturbation methods. Robustness is then evaluated using Fast Gradient Sign Method (FGSM), Basic Iterative Method (BIM), and Projected Gradient Descent (PGD) attacks. Root Mean Squared Error (RMSE) degradation is used as the evaluation metric. Every attack configuration is repeated across 30 randomized iterations to provide comparisons that are consistent with common Monte Carlo evaluation practices. Experiments are conducted on two real world datasets: the Jena Climate dataset and Electricity Load Diagrams dataset. Results show that both ensembles improve robustness relative to the base model. The novel perturbation strategy achieves competitive or superior performance under BIM and PGD across most perturbation budgets