Search Results (288)

Nowadays, a pedestrian navigation system using a smartphone has become popular as a useful tool to reach an unknown destination. When the destination is the office of a person, a detailed map information is necessary on the target area such as the room number and location inside the building. The information can be collected from various sources including Google maps, websites for the building, and images of signs. In this paper, we propose a map information collection tool for a pedestrian navigation system. To improve the accuracy and completeness of information, it works with the four steps: (1) a user captures building and room images manually, (2) an OCR software using Google ML Kit v2 processes them to extract the sign information from images, (3) web scraping using Scrapy (v2.11.0) and crawling with Apache Nutch (v1.19) software collects additional details such as room numbers, facilities, and occupants from relevant websites, and (4) the collected data is stored in the database to be integrated with a pedestrian navigation system. For evaluations of the proposed tool, the map information was collected for 10 buildings at Okayama University, Japan, a representative environment combining complex indoor layouts (e.g., interconnected corridors, multi-floor facilities) and high pedestrian traffic, which are critical for testing real-world navigation challenges. The collected data is assessed in completeness and effectiveness. A university campus was selected as it presents a complex indoor and outdoor environment that can be ideal for testing pedestrian navigations in real-world scenarios. With the obtained map information, 10 users used the navigation system to successfully reach destinations. The System Usability Scale (SUS) results through a questionnaire confirms the high usability. Full article

Background: As e-commerce expands and delivery services diversifies, understanding the factors that shape consumer preferences becomes critical to designing efficient and sustainable urban logistics. This study examines how perceived walkability influences consumers’ preferences for shopping channels (in-store or online) and delivery methods (home delivery versus pickup points). Method: The analysis is based on structural equation modeling and utilizes survey data collected from 444 residents of Belo Horizonte, Brazil. Results: The findings emphasize the importance of walkability in supporting weekday store visits, encouraging pickup for online purchases and fostering complementarity between different modes of purchase and delivery services. Perceived walkability positively affects the preference to buy in physical stores and increases the likelihood of using pickup points. Educated men, particularly those living in walkable areas, are the most likely to adopt pickup services. In contrast, affluent individuals and women are less likely to forgo home delivery in favor of pickup points. Conclusions: The results highlight the role of perceived walkability in encouraging in-person pickup as a sustainable alternative to home delivery, providing practical guidance for retailers, urban planners, and logistics firms seeking to align consumer convenience with sustainable delivery strategies. Full article

An Identity Management System (IDMS) is responsible for managing and organizing identities and credentials exchanged between users, Identity Providers (IDPs), and Service Providers (SPs). The primary goal of IDMS is to ensure the confidentiality and privacy of users’ personal data. Traditional IDMS relies on a third party to store user information and authenticate the user. However, this approach poses threats to user privacy and increases the risk of single point of failure (SPOF), user tracking, and data unavailability. In contrast, decentralized IDMSs that use blockchain technology offer potential solutions to these issues as they offer powerful features including immutability, transparency, anonymity, and decentralization. Despite its advantages, blockchain technology also suffers from limitations related to performance, third-party control, weak authentication, and data leakages. Furthermore, some blockchain-based IDMSs still exhibit centralization issues, which can compromise user privacy and create SPOF risks. This study proposes a decentralized IDMS that leverages blockchain and smart contract technologies to address the shortcomings of traditional IDMSs. The proposed system also utilizes the Interplanetary file system (IPFS) to enhance the scalability and performance by reducing the on-chain storage load. Additionally, the proposed IDMS employs the Elliptic Curve Integrated Encryption Scheme (ECIES) to provide an extra layer of security to protect users’ sensitive information while improving the performance of the systems’ transactions. Security analysis and experimental results demonstrated that the proposed IDMS offers significant security and performance advantages compared to its counterparts. Full article

Carsharing has long supported trip purposes typically made by private vehicles, with grocery shopping especially benefiting from the carrying capacity of a personal vehicle. BlueLA is a one-way, station-based electric vehicle (EV) carsharing service in Los Angeles aimed at improving access in low-income neighborhoods. We hypothesize that BlueLA improves grocery access for underserved households by increasing their spatial-temporal reach to diverse grocery store types. We test two hypotheses: (1) accessibility from BlueLA stations to grocery stores varies by store type, traffic conditions, and departure times; and (2) Standard (general population) and Community (low-income) members differ in perceived grocery access and station usage. Using a mixed-methods approach, we integrate walking and driving isochrones, store data (n = 5888), trip activity data (n = 59,112), and survey responses (n = 215). Grocery shopping was a key trip purpose, with 69% of Community and 61% of Standard members reporting this use. Late-night grocery access is mostly limited to convenience stores, while roundtrips to full-service stores range from 55 to 100 min and cost USD 12 to USD 20. Survey data show that 84% of Community and 71% of Standard members reported improved grocery access. The findings highlight the importance of trip timing and the potential for carsharing and retail strategies to improve food access. Full article

Data generated by Internet of Things devices enable the design of new business models and services, improving user experience and satisfaction. This data also serve as an essential information source for many fields, including disaster management, bio-surveillance, smart cities, and smart health. However, personal data are also collected in this context, introducing new challenges concerning data privacy protection, such as profiling, localization and tracking, linkage, and identification. This dilemma is further complicated by the “privacy paradox”, where users compromise privacy for service convenience. Hence, this paper reviews the literature on data privacy in the IoT, particularly emphasizing Personal Data Store (PDS)-based approaches as a promising class of user-centric solutions. PDS represents a user-centric approach to decentralizing data management, enhancing privacy by granting individuals control over their data. Addressing privacy solutions involves a triad of user privacy awareness, technology support, and ways to regulate data processing. Our discussion aims to advance the understanding of IoT privacy issues while emphasizing the potential of PDS to balance privacy protection and service delivery. Full article

Product engagement detection in retail environments is critical for understanding customer preferences through nonverbal cues such as gaze and hand movements. This study presents a system leveraging a 360-degree top-view fisheye camera combined with two perspective cameras, the only sensors required for deployment, effectively capturing subtle interactions even under occlusion or distant camera setups. Unlike conventional image-based gaze estimation methods that are sensitive to background variations and require capturing a person’s full appearance, raising privacy concerns, our approach utilizes a novel Transformer-based encoder operating directly on 3D skeletal keypoints. This innovation significantly reduces privacy risks by avoiding personal appearance data and benefits from ongoing advancements in accurate skeleton estimation techniques. Experimental evaluation in a simulated retail environment demonstrates that our method effectively identifies critical gaze-object and hand-object interactions, reliably detecting customer engagement prior to product selection. Despite yielding slightly higher mean angular errors in gaze estimation compared to a recent image-based method, the Transformer-based model achieves comparable performance in gaze-object detection. Its robustness, generalizability, and inherent privacy preservation make it particularly suitable for deployment in practical retail scenarios such as convenience stores, supermarkets, and shopping malls, highlighting its superiority in real-world applicability. Full article

Coxiella burnetii is a significant infectious pathogen that causes Q fever. Q fever is thought to be uncommon in the US and most human cases are believed to occur in agricultural livestock workers. However, the extent of US community exposure to C. burnetii is not known with certainty. Using nationally representative 2003–2004 US National Health and Nutrition Examination Survey serologic, demographic, and occupational history data, the magnitude of US adult general population exposure to C. burnetii, excluding agricultural-sector workers, was estimated. Exposure was defined as positive serum IgG antibodies in an immunofluorescence assay (e.g., current or past infection). A total of 3.0% (95% CI: 2.0–4.4) of the US population met the criteria for C. burnetii exposure, representing some 6.2 million persons. Overall, 86.9% (95% CI: 75.5–98.4) of the seropositive persons had no lifetime history of work in the agricultural sector (5.5 million persons). This was consistently true across all US demographic groups: aged 20–59 years, 87.3%; aged 60+ years, 85.7%; men, 86.1%; women, 87.6%; non-Hispanic Whites, 82%; non-Hispanic Blacks, 95.8%; Mexican Americans, 89.4%; immigrants from Mexico, 83.5%; and other immigrants, 96.8%. As a proportion of C. burnetii infections result in acute Q fever and chronic Q fever conveys significant mortality, the community-level risks to the general public may be significant. It is recommended that a 6-year sample of the most recent NHANES stored sera be analyzed to determine the current community C. burnetii exposure rates. Also, analyzing an additional 2005–2008 stored sera sample would provide an opportunity to assess the time trends and long-term health impacts. Full article

Recently, the increasing frequency of cybersecurity incidents has raised concerns about communication security and personal privacy. In a zero-trust network environment, it is critically important to protect communication content and ensure that it is not intercepted, recorded, or stored without proper authorization. End-to-end encryption (E2EE) is a reliable solution for this purpose. The COVID-19 pandemic has accelerated the adoption of remote work and virtual meetings, making the security of voice conferences a critical issue. This study aims to explore the application of end-to-end encryption technology in voice conferences. We designed and implemented an end-to-end encrypted voice conferencing system based on terminal-side mixing to ensure security while also being applicable in low-bandwidth network environments. The developed system effectively prevented man-in-the-middle attacks and data wiretaps, while maintaining high performance and low latency. It can be used in low-bandwidth scenarios such as satellite networks. The end-to-end encryption technology, when combined with terminal-side voice mixing, significantly enhances the security and usability of voice conferences as a new solution for secure communication in the future. Full article

The protection of personally identifiable information (PII) is being increasingly demanded by customers and governments via data protection regulations. Private and public organizations store and exchange through the Internet a large amount of data that include the personal information of users, employees, and customers. While discovering PII from a large unstructured text corpus is still challenging, a lot of research work has focused on identifying methods and tools for the detection of PII in real-time scenarios and the ability to discover data exfiltration attacks. In those research attempts, natural language processing (NLP)-based schemas are widely adopted. Our work combines NLP with deep learning to identify PII in unstructured texts. NLP is used to extract semantic information and the syntactic structure of the text. This information is then processed by a pre-trained Bidirectional Encoder Representations from Transformers (BERT) algorithm. We achieved high performance in detecting PII, reaching an accuracy of 99.558%. This represents an improvement of 7.47 percentage points over the current state-of-the-art model that we analyzed. However, the experimental results show that there is still room for improvement to obtain better accuracy in detecting PII, including working on a new, balanced, and higher-quality training dataset for pre-trained models. Our study contributions encourage researchers to enhance NLP-based PII detection models and practitioners to transform those models into privacy detection tools to be deployed in security operation centers. Full article

Background. Inflammatory bowel disease (IBD), encompassing Crohn’s disease (CD) and ulcerative colitis (UC), is a complex inflammatory condition affecting the intestinal tract. Currently, immune-modulating treatments are inadequate for 30–50% of patients and often cause significant side effects, highlighting the urgent need for a personalized medicine approach. Real-world data and archived gut biological material from clinical repositories could be a resource for identifying new drug candidates and biomarkers. This study assesses the extent of stored formalin-fixed, paraffin-embedded (FFPE) gut biopsies from patients with IBD that could be leveraged for research efforts. Methods. Data from the Danish National Patient Register and the Danish Pathology Register were used to construct a cohort of patients diagnosed with IBD between 1 January 2005, and 30 June 2013, and followed for five years. Results. Among 14,512 IBD patients, 13,936 (96%) had at least one biopsy visit within five years after their initial diagnosis (CD 94%, UC 97%), and 13,598 (94%) had their first biopsy visit as part of the diagnostic process. Biopsies were taken from the colon (82%) or multiple locations (46%). Patients with severe disease had more biopsy visits than those with non-severe disease (IBD 3.3 vs. 2.0 visits, CD 2.9 vs. 1.9 visits, UC 3.6 vs. 2.0 visits). Conclusions. Thus, the vast majority of patients with IBD have biopsies taken. These findings demonstrate the feasibility and applicability of combining real-world data and archived gut biopsies for research, highlighting it as a valuable but underutilized resource for identifying new drug candidates and biomarkers, with huge potential for enhancing personalized medicine within IBD for the benefit of patients and society. Full article

An ultralow-power, high-performance online-learning and anomaly-detection system has been developed for edge security applications. Designed to support personalized learning without relying on cloud data processing, the system employs sample-wise learning, eliminating the need for storing entire datasets for training. Built using memristor-based analog neuromorphic and in-memory computing techniques, the system integrates two unsupervised autoencoder neural networks—one utilizing optimized crossbar weights and the other performing real-time learning to detect novel intrusions. Threshold optimization and anomaly detection are achieved through a fully analog Euclidean Distance (ED) computation circuit, eliminating the need for floating-point processing units. The system demonstrates 87% anomaly-detection accuracy; achieves a performance of 16.1 GOPS—774× faster than the ASUS Tinker Board edge processor; and delivers an energy efficiency of 783 GOPS/W, consuming only 20.5 mW during anomaly detection. Full article

As the use of personal mobility (PM) devices continues to rise, regulatory violations have become more frequent, highlighting the need for technological solutions to ensure efficient enforcement. This study addresses these challenges by proposing an AI-based enforcement platform. The system integrates the You Only Look Once version 5 (YOLOv5) object detection model, a deep-learning-based framework, with Global Positioning System (GPS) location data, Raspberry Pi 5, and Amazon Web Services (AWS) for data processing and web-based implementation. The YOLOv5 model was deployed in two configurations: one for detecting electric scooter usage and another for identifying legal violations. The system utilized AWS Relational Database Service (RDS), Simple Storage Service (S3), and Elastic Compute Cloud (EC2) to store violation records and host web applications. The detection performance was evaluated using mean average precision (mAP) metrics. The electric scooter detection model achieved mAP50 and mAP50-95 scores of 99.5 and 99.457, respectively. Meanwhile, the legal violation detection model attained mAP50 and mAP50-95 scores of 99.5 and 81.813, indicating relatively lower accuracy for fine-grained violation detection. This study presents a practical technological platform for monitoring regulatory compliance and automating fine enforcement for shared electric scooters. Future improvements in object detection accuracy and real-time processing capabilities are expected to enhance the system’s overall reliability. Full article

Empathetic and coherent responses are critical in automated chatbot-facilitated psychotherapy. This study addresses the challenge of enhancing the emotional and contextual understanding of large language models (LLMs) in psychiatric applications. We introduce Emotion-Aware Embedding Fusion, a novel framework integrating hierarchical fusion and attention mechanisms to prioritize semantic and emotional features in therapy transcripts. Our approach combines multiple emotion lexicons, including NRC Emotion Lexicon, VADER, WordNet, and SentiWordNet, with state-of-the-art LLMs such as Flan-T5, Llama 2, DeepSeek-R1, and ChatGPT 4. Therapy session transcripts, comprising over 2000 samples, are segmented into hierarchical levels (word, sentence, and session) using neural networks, while hierarchical fusion combines these features with pooling techniques to refine emotional representations. Attention mechanisms, including multi-head self-attention and cross-attention, further prioritize emotional and contextual features, enabling the temporal modeling of emotional shifts across sessions. The processed embeddings, computed using BERT, GPT-3, and RoBERTa, are stored in the Facebook AI similarity search vector database, which enables efficient similarity search and clustering across dense vector spaces. Upon user queries, relevant segments are retrieved and provided as context to LLMs, enhancing their ability to generate empathetic and contextually relevant responses. The proposed framework is evaluated across multiple practical use cases to demonstrate real-world applicability, including AI-driven therapy chatbots. The system can be integrated into existing mental health platforms to generate personalized responses based on retrieved therapy session data. The experimental results show that our framework enhances empathy, coherence, informativeness, and fluency, surpassing baseline models while improving LLMs’ emotional intelligence and contextual adaptability for psychotherapy. Full article

In recent years, advances in diagnosing and classifying diseases using machine learning (ML) have grown exponentially. However, due to the many privacy regulations regarding personal data, pooling together data from multiple sources and storing them in a single (centralized) location for traditional ML model training are often infeasible. Federated learning (FL), a collaborative learning paradigm, can sidestep this major pitfall by creating a global ML model that is trained by aggregating model weights from individual models that are separately trained on their own data silos, therefore avoiding most data privacy concerns. This study addresses the centralized data issue with FL by applying a novel DataWeightedFed architectural approach for effective fundus disease diagnosis from ophthalmic images. It includes a novel method for aggregating model weights by comparing the size of each model’s data and taking a dynamically weighted average of all the model’s weights. Experimental results showed a small average 1.85% loss in accuracy when training using FL compared to centralized ML model systems, a nearly 92% improvement over the conventional 55% accuracy loss. The obtained results demonstrate that this study’s FL architecture can maximize both privacy preservation and accuracy for ML in fundus disease diagnosis and provide a secure, collaborative ML model training solution within the eye healthcare space. Full article

Standards describing the test procedures recommended to investigate the shielding effectiveness of enclosures have two major issues: they generally prescribe the assessment of the electromagnetic field of empty cavities, and they do not deal with very small enclosures. However, the dimensions of some very common shielded apparatus are smaller than those considered in the standards and the electromagnetic field distribution inside the shielded structure is strongly affected by the enclosure content. In this paper, both issues have been investigated for two components commonly used in medium voltage/low voltage (MV/LV) substations: a mini personal computer used to store, process, and transmit relevant data on the status of the electric network, with these aspects being essential in smart grids, and an electronic relay which is ubiquitous in MV/LV substations. Both components are partially contained in a metallic enclosure which provides a certain amount of electromagnetic shielding against external interferences. It is observed that an electrostatic discharge may cause a failure and/or a loss of data, requiring an improvement of shielding characteristics or a wise choice of the positions where the most sensitive devices are installed inside the enclosure. Since the dimensions of very small enclosures, fully occupied by their internal components, do not allow for the insertion of sensors inside the protected volume, numerical analysis is considered as the only way for the appraisal of the effects induced by a typical source of interference, such as an electrostatic discharge. Full article