Search

In heterogeneous networks where macro cells and metro cells use the same frequency band to communicate with their respective users, the major problem limiting performance is the interference caused by macro cells to metro cells. The information theoretic result known as dirty paper coding provides a way to address this problem and significantly improve the performance of heterogeneous networks with co-channel deployment. In this paper, we show how a simple dirty paper coding scheme employing Tomlinson-Harashima pre-coding with partial interference pre-subtraction can be employed by metro cells to mitigate the interference caused by macro cells. A performance study included in this paper shows that the proposed dirty paper coding scheme can lead to significant improvement in user rate statistics.

Zone identification is a topic in the area of text mining which helps researchers be benefited by the content of scientific papers in a satisfactory manner. The major aim of zone identification is to classify the sentences of scientific texts into some predefined zone categories which can be useful for summarization as well as information extraction. In this paper, we propose a two-level approach to zone identification within which the first level is in charge of classifying the sentences in a given paper based on some semantic and lexical features. In this respect, several machine learning algorithms such as Simple Logistics, Logistic Model Trees and Sequential Minimal Optimization are applied. The second level is responsible for applying fusion to the classification results obtained for consecutive sentences of the first level in order to make the final decision. The proposed method is evaluated on ART and DRI corpora as two well-known data sets. Results obtained for the accuracy of zone identification for these corpora are respectively 65.75% and 84.15%, which seem to be quite promising compared to those obtained by previous approaches.

Researchers from academia and the corporate-sector rely on scholarly digital libraries to access articles. Attackers take advantage of innocent users who consider the articles' files safe and thus open PDF-files with little concern. In addition, researchers consider scholarly libraries a reliable, trusted, and untainted corpus of papers. For these reasons, scholarly digital libraries are an attractive-target and inadvertently support the proliferation of cyber-attacks launched via malicious PDF-files. In this study, we present related vulnerabilities and malware distribution approaches that exploit the vulnerabilities of scholarly digital libraries. We evaluated over two-million scholarly papers in the CiteSeerX library and found the library to be contaminated with a surprisingly large number (0.3-2%) of malicious PDF documents (over 55% were crawled from the IPs of US-universities). We developed a two layered detection framework aimed at enhancing the detection of malicious PDF documents, Sec-Lib, which offers a security solution for large digital libraries. Sec-Lib includes a deterministic layer for detecting known malware, and a machine learning based layer for detecting unknown malware. Our evaluation showed that scholarly digital libraries can detect 96.9% of malware with Sec-Lib, while minimizing the number of PDF-files requiring labeling, and thus reducing the manual inspection efforts of security-experts by 98%.

Aiming at the problem of easy tampering and difficult integrity authentication of paper text documents, this paper proposes a robust content authentication method for printed documents based on text watermarking scheme resisting print-and-scan attack. Firstly, an authentication watermark signal sequence related to content of text document is generated based on the Logistic chaotic map model; then, the authentication watermark signal sequence is embedded into printed paper document by using a robust text watermarking scheme; finally, the watermark information is extracted from scanned image of paper document, and compared with the authentication watermark information calculated in real time by the text document content obtained by OCR technology, thereby performing content integrity authentication of the paper text documents. Experimental results show that our method can achieve the robust content integrity authentication of paper text documents, and can also accurately locate the tampering position. In addition, the document after embedding the watermark information has a good visual effect, and the text watermarking scheme has a large information capacity.

Many home IoT devices suffer from poor security design and confusing interfaces, lowering the bar for successful cyberattacks. A popular approach to identify compromised IoT devices is network-based detection, in which network traffic is analyzed to fingerprint and identify such devices. However, while several network-based techniques for identifying misbehaving devices have been proposed, the role of the user in remediating IoT security incidents has been conspicuously overlooked.

Containers are a mainstream virtualization technique commonly used to run stateful workloads over persistent storage. In multi-tenant hosts with high utilization, resource contention at the system kernel often leads to inefficient handling of the container I/O. Assuming a distributed storage architecture for scalability, resource sharing is particularly problematic at the client hosts serving the applications of competing tenants. Although increasing the scalability of a system kernel can improve resource efficiency, it is highly challenging to refactor the kernel for fair access to system services. As a realistic alternative, we isolate the storage I/O paths of different tenants by serving them with distinct clients running at user level. We introduce the Danaus client architecture to let each tenant access the container root and application filesystems over a private host path. We developed a Danaus prototype that integrates a union filesystem with a Ceph distributed filesystem client and a configurable shared cache. Across different host configurations, workloads and systems, Danaus achieves improved performance stability because it handles I/O with reserved per-tenant resources and avoids intensive kernel locking. Danaus offers up to 14.4x higher throughput than a popular kernel-based client under conditions of I/O contention. In comparison to a FUSE-based user-level client, Danaus also reduces by 14.2x the time to start 256 high-performance webservers. Based on our extensive experience from building and evaluating Danaus, we share several valuable lessons that we learned about resource contention, file management, service separation and performance stability.

Function-as-a-Service (FaaS) has raised a growing interest in how to "tame" serverless to enable domain-specific use cases such as data-intensive applications and machine learning (ML), to name a few. Recently, several systems have been implemented for training ML models. Certainly, these research articles are significant steps in the correct direction. However, they do not completely answer the nagging question of when serverless ML training can be more cost-effective compared to traditional "serverful" computing. To help in this task, we propose MLLess, a FaaS-based ML training prototype built atop IBM Cloud Functions. To boost cost-efficiency, MLLess implements two key optimizations: a significance filter and a scale-in auto-tuner, and leverages them to specialize model training to the FaaS model. Our results certify that MLLess can be 15X faster than serverful ML systems [24] at a lower cost for ML models (such as sparse logistic regression and matrix factorization) that exhibit fast convergence.

Students frequently utilize digital devices and applications to assist studying. Past research has yielded mixed results on their effectiveness, and scant research has compared portable technologies with each other or focused on smartphones specifically. We compared college students’ learning of vocabulary on paper, laptop, and smartphone. Students were randomly assigned to one platform for reviewing vocabulary flashcards, and then they completed a memory quiz and answered questions regarding perceptions of these technologies. Results showed equivalency in students’ learning across platforms, with the only exception being that students spent longer reviewing paper flashcards. Self-reported cognitive load and satisfaction were equal across platforms. Quantitative and qualitative data also suggested that, though students quickly label both laptops and smartphones as helpful for education, they see more educational value in laptops. This research suggests that smartphones present new ways of learning, but care must be taken with how smartphones are utilized in educational settings.

A database of historic paper properties was developed using 729 samples of European origin (1350–1990), analysed for acidity, degree or polymerisation (DP), molecular weight of cellulose, grammage, tensile strength, as well as contents of ash, aluminium, carbonyl groups, rosin, protein, lignin and fibre furnish. Using Spearman’s rank correlation coefficient and principal component analysis, the data were examined with respect to methods of manufacture, as well as chemical stability of paper. Novel patterns emerged related to loss of DP and accumulation of carbonyl groups and acidity with time and the role of lignin and rosin, as well as rate of degradation (k = 10−5 year−1) at room conditions. In-depth understanding of long-term degradation of lignin and rosin is needed to better understand the relationships between composition and degradation of historic paper. This study highlights the importance of mining significant volumes of analytical data, and its variability, obtained from real historic objects.

Paper writing with the existing writing format causing excessive paper usage. Excessive paper usage could lead to environmental destruction. Therefore, it is necessary to find an alternative in the writing procedure by doing an experimental design based on student reading comprehensive. This study uses Completely Randomized Factorial Experimental Design. After analysis of variance and the multiple comparisons test done, it results a new writing format alternative that use less paper than the existing writing procedure. In other hand, it results greater student reading comprehensive than the existing writing procedure. With the results, the application of the alternative procedure can save the paper usage and make student easier to understand the paper .

Internet of Things (IoT) devices are increasingly being found in civilian and military contexts, ranging from smart cities and smart grids to Internet-of-Medical-Things, Internet-of-Vehicles, Internet-of-Military-Things, Internet-of-Battlefield-Things, etc. In this paper, we survey articles presenting IoT security solutions published in English since January 2016. We make a number of observations, including the lack of publicly available IoT datasets that can be used by the research and practitioner communities. Given the potentially sensitive nature of IoT datasets, there is a need to develop a standard for sharing IoT datasets among the research and practitioner communities and other relevant stakeholders. Thus, we posit the potential for blockchain technology in facilitating secure sharing of IoT datasets (e.g., using blockchain to ensure the integrity of shared datasets) and securing IoT systems, before presenting two conceptual blockchain-based approaches. We then conclude this paper with nine potential research questions. Keywords: Blockchain, Blockchain security, Collaborative security, Internet-of-military things, IoT dataset, IoT self-healing, IoT security, Intrusion-prevention system, Predictive IoT security, Predictive security

Thermal papers, commonly used for printed receipts or lottery tickets, are omnipresent in our everyday life. They are regarded as semi-porous substrates, and yet can be critical to analyze when looking for latent fingermarks due to their thermosensibility. The aim of this study was to investigate a development sequence that would better combine the adequate detection techniques in order to maximize the chances to develop latent fingermarks left on these substrates. Different methods of development have been compared on test substrates: black magnetic powder, Lumicyano™, thermal development, ninhydrin and 1,2-indanedione/ZnCl2. Whitening stages and thermal development have been focused on, tested and optimized. The results of these preliminary tests enabled the study of three development sequences. They have subsequently been compared to the one currently used in the Gendarmerie's laboratories and the best results have been provided during pseudo-operational comparative trials by one of these sequences, consisting in 6 stages.

One of the benefits of Internet of Vehicles (IoV) is improved traffic safety and efficiency, for example due to the capability to share vehicular messages in real-time. While most of the vehicular messages only need to be shared by nearby vehicles, some messages (e.g., announcement messages) may need to be more broadly distributed, for example to vehicles in a wider region. Finding a single trusted entity to store and distribute such messages can be challenging, and vehicles may not be inclined to participate (e.g., generation and distribution of announcement messages) unless they can benefit from such participation. In addition, achieving both security and privacy can be challenging. In this paper, we propose a blockchain based secure data sharing system to address the above challenges in an IoV setting. Specifically, in our system, announcement messages are stored using blockchain. To encourage/incentivize participation, vehicles that faithfully broadcast the announcement messages and/or contribute to the block generation will be rewarded by some cryptocurrency. Our system is also designed to be privacy-preserving and realizes both priori and posteriori countermeasures.

School psychologists are increasingly tasked with assessing, supporting, and intervening with dual language learner (DLL) students, their teachers, and their families. Understanding the assets of bilingualism along with multiple issues associated with comprehensive practice with DLL has become paramount. Currently, practitioners often lack the depth in knowledge of cultural variables, dual language acquisition, knowledge of programs to effectively serve DLL students, bilingual assessment, and research and evidence-based practice to serve DLLs competently, as well as depth in second language competence. This white paper, endorsed by the School Psychology Educators of California (SPEC) and the California Association of School Psychologists (CASP), outlines areas of competence deemed to be essential to all psychologists, as well as additional areas of competence for practitioners identifying themselves bilingual school psychologists.

These data were used to build an object detection model to locate Relict Charcoal Hearths (RCH) as described in the paper “When Computers Dream of Charcoal: Using Deep Learning, Open Tools and Open Data to Identify Relict Charcoal Hearths in and around State Game Lands in Pennsylvania” [1]. This is the first grouping of data for the paper above. The second grouping is also available in this journal, see “Object detection model, image data and results from the “When Computers Dream of Charcoal: Using Deep Learning, Open Tools and Open Data to Identify Relict Charcoal Hearths in and around State Game Lands in Pennsylvania” paper”. These files consist of georeferenced Digital Elevation Model, Hillshade and Slope files derived from LiDAR for the State Game Lands (SGL). Included is a Shapefile and GeoJSON of State Game Land borders as well as the program used for downloading the LiDAR files. These data are stored on Zenodo.org .

This companion paper is to support the replication of paper "Campus3D: A Photogrammetry Point Cloud Benchmark for Outdoor Scene Hierarchical Understanding", which was presented at ACM Multimedia 2020. The supported paper's main purpose was to provide a photogrammetry point cloud-based dataset with hierarchical multilabels to facilitate the area of 3D deep learning. Based on this provided dataset and source code, in this work, we build a complete package to reimplement the proposed methods and experiments (i.e., the hierarchical learning framework and the benchmarks of the hierarchical semantic segmentation task). Specifically, this paper contains the technical details of the package, including file structure, dataset preparation, installation package, and the conduction of the experiment. We also present the replicated experiment results and indicate our contributions to the original implementation.

In this companion paper, we provide the details of the reproducibility artifacts of the paper "Visual Relation of Interest Detection" presented at MM'20. Visual Relation of Interest Detection (VROID) aims to detect visual relations that are important for conveying the main content of an image. In this paper, we explain the file structure of the source code and publish the details of our ViROI dataset, which can be used to retrain the model with custom parameters. We also detail the scripts for component analysis and comparison with other methods and list the parameters that can be modified for custom training and inference.

In this paper we reproduce experimental results presented in our earlier work titled "Describing Subjective Experiment Consistency by $p$-Value P-P Plot" that was presented in the course of the 28th ACM International Conference on Multimedia. The paper aims at verifying the soundness of our prior results and helping others understand our software framework. We present artifacts that help reproduce tables, figures and all the data derived from raw subjective responses that were included in our earlier work. Using the artifacts we show that our results are reproducible. We invite everyone to use our software framework for subjective responses analyses going beyond reproducibility efforts., Comment: Please refer to the original publication: https://dl.acm.org/doi/10.1145/3474085.3477935 Related paper: https://dl.acm.org/doi/abs/10.1145/3394171.3413749 or arXiv:2009.13372

Peran teknologi informasi bagi perusahaan mempunyai peranan penting dan dapat menjadi pusat strategi bisnis untuk memperoleh keunggulan bersaing. Saat ini teknologi informasi sudah menjadi kebutuhan dasar bagi setiap perusahaan. Perusahaan yang ingin dikenal secara luas maupun perusahaan yang ingin menjaga hubungan dengan pelanggannya perlu membuat situs/website yang berisi informasi profil perusahaan maupun jasa/produk yang diberikan. Setiap website pasti memiliki sebuah user interface yang berbeda-beda. Klinik Sehat sama seperti klinik pada umumnya yang memberikan pelayanan yang efektif dan efisien. Untuk meningkatkan pelayanan klinik, pihak klinik ingin membuat website yang dapat digunakan pengguna untuk mudah melihat aktifitas yang sedang berlangsung di klinik. Dengan membuat website klinik, tampilan antarmuka (interface) berperan penting dalam pembuatan website. Interface merupakan hal yang di lihat pertama kali pada saat pengguna membuka website Klinik Sehat. Interface menjadi patokan. Untuk membuat interface yang bersifat user friendly pihak kinik membuat interface dengan metode paper prototype. Paper prototype dapat membantu membuat interface pada awal pengembangannya. Hasil yang di dapat pada paper ini adalah metode paper prototype cocok digunakan untuk membuat tampilan user interface pada website seperti Klinik Sehat, tampilan user interface yang user friendly bagi para pengguna website, Pengguna dapat melakukan ujicoba antarmuka dengan aksi-aksi yang nyata dan juga dapat menentukan sendiri aksi atau tugas apa saja yang bisa dipahami dan dilakukan ketika menggunakan aplikasi/website tersebut secara intuitif.

This chapter focuses on cost control by analyzing the intelligent control system of waste paper deinking production line, and finally designs a solution. In the aspect of control index, machine learning has designed a solution that pays less attention to the benefit of a single control loop and focuses on maximizing the overall benefit of the production line. In the aspect of control demand, according to the complexity of waste paper deinking production line, a distributed control solution is designed Finally, according to the dynamic fuzziness in machine learning system, the basic concept of dynamic fuzzy machine learning is constructed by using dynamic fuzzy set The dynamic fuzzy machine learning model is described and the related learning algorithm is given. Finally, a simulation example is given, and satisfactory results are obtained by comparing with the elastic BP algorithm RPROP of neural network.

This work is centered on the paper making process in an industrial paper mill. There exist plenty of sub-processes in a paper mill for transformation of raw material to finished product. A particular process that is taken up in this research is paper basis weight control process, which is very important for good quality paper production. But the controller design for this process is very challenging due to the involvement of severe dead time value, which in addition to delaying the control output, makes the control system unstable. So, two different approaches are used here, in MATLAB, for controller design, one is the dead time approximation using Pade approximation technique and the other is dead time compensation using Smith predictor technique. Hence, the performance of the designed control systems is compared.

When students believe they can get smarter, they understand that effort makes them stronger. Therefore, they put in extra time and effort, and that leads to higher achievement. Besides that, the arrangement of question papers and answer sheets process for a high number of students takes a long time. In this paper, the sheet exam has been proposed to change into an electric exam (E-exam). This system depended on the client-server framework to convert the traditional exam environment into an electronic exam, providing multiple and different questions at the same time. In additions, kept the grades of students and providing them automatically. In this paper, Visual Studio and Microsoft SQL server have been proposed to develop the electronic exam. The questionnaire is made by distributed it among 30 students, and after the data analysis, the results have been collected have been represented a response rate of 100%. It is recommended that, to take electronic exams and e-learning for students periodically. To improve their performance through continuous training on computerized exams and to increase the student’s efficiency in this type of learning.

We introduce a comprehensive benchmark for local features and robust estimation algorithms, focusing on the downstream task -- the accuracy of the reconstructed camera pose -- as our primary metric. Our pipeline's modular structure allows easy integration, configuration, and combination of different methods and heuristics. This is demonstrated by embedding dozens of popular algorithms and evaluating them, from seminal works to the cutting edge of machine learning research. We show that with proper settings, classical solutions may still outperform the perceived state of the art. Besides establishing the actual state of the art, the conducted experiments reveal unexpected properties of Structure from Motion (SfM) pipelines that can help improve their performance, for both algorithmic and learned methods. Data and code are online https://github.com/vcg-uvic/image-matching-benchmark, providing an easy-to-use and flexible framework for the benchmarking of local features and robust estimation methods, both alongside and against top-performing methods. This work provides a basis for the Image Matching Challenge https://vision.uvic.ca/image-matching-challenge., Comment: Added: KeyNet-SOSNet, AffNet-HardNet, TFeat, MKD from kornia

Purpose of the study: This study aimed to utilize the usage of App in Public Speaking course to ease and engage students of Office Management & Technology (OMT) in their learning experience by providing the simplified version of speech draft development. Methodology: The development of the App divided into two divisions, (a) the App content and, (b) the App design and concept. The content of the App was from the main textbook, simplified to serve the needs of OMT students, whereas the Apps’ design and concept were based on interactive design and waterfall models. Analysis, Design, Development, Implementation, and Evaluation (ADDIE) and Focus Group were also included. Main Findings: The Infusion Speech Skeleton App is projected to be a vital platform for students to improve their understanding of the course. The App is also expected to improve students’ performance particularly those who have trouble comprehending speech draft development. Applications of this study: Because the App is mainly developed with OMT students in the loop, most of the settings will be in OMT students’ favors. However, the Apps also can be beneficial to those taking literature, and other communications studies for basic reference. Novelty/Originality of this study: The App usage in learning was introduced in many courses, however, to our best knowledge this is the first attempt of App development and implementation in the course of public speaking, focusing on the need for non-communication students, OMT students.

This study investigated the performance of four widely used data-collection designs in detecting test-mode effects (i.e., computer-based versus paper-based testing). The experimental conditions inc...

This article introduces Turing’s idea of a “paper machine” to identify and understand one important mode of clinical research in the modern hospital, how that research worked, and how offic...

Usually, scientific research begins with the collection of data in which online social media tools can be some of the most rewarding and informative resources. The extensive measure of accessible information pulls in users from undergraduate students to postdoc. The search for scientific themes has popularized due to the availability of abundant publications that resides in scientific social networks such as Mendeley, ResearchGate etc. Articles are published on these media inform of text for knowledge dissemination, scientific support, research, updates etc, and are frequently uploaded after its publication in a proceedings or journal. In this sense, data collected from database often contains high noise and its analysis can be treated as a characterization undertaking as it groups the introduction of a content into either good or bad. In this text, we present quantitative and qualitative analysis of papers popularity in Mendeley repository by using naive Bayes Classifier.

Recent advances in telecommunication technology and video conferencing systems have opened a new avenue for surgical instruction called “surgical telementoring.” This report from the Technology Working Group of the SAGES Project 6 Summit reviews the telementoring technology that currently exists and proposes recommendations for minimum technology requirements and future technology development. While also providing insight in regulatory considerations, this review offers what prospective surgical telementoring participants need to know about the underlying technology with a specific focus on safety, reliability, transmission quality, ease of use, and cost. Content experts from around the world, in minimally invasive surgery, surgical mentoring and telementoring, surgical education, business development, healthcare innovation, and regulation were invited to attend a 2-day summit in Los Angeles, USA to outline the current state of surgical telementoring and chart the challenges and opportunities going forward. This article summarizes the discussion, conclusions, and recommendation of the technology group with regard to telementoring technology. This article reviews the technical requirements which can be divided into the following categories: (1) safety, (2) reliability, (3) transmission quality, (4) ease of use, and (5) cost. Telementoring applications are technology driven. Given the pace of change of technology, guiding principles in technology design and selection are warranted (Table 4). Telementoring technologies require two basic components, video capturing and display devices at the transmitting and receiving end, and a telecommunication link between them. Many additional features can be added to this basic setup including multiple cameras or video sources, remote camera zoom and pan, recording and storage of videos and images, and telestration capabilities to mention just a few. In general, the cost of these technologies is feature driven. The education framework for each specific application should determine the need for these features (Schlachta in Surg Endosc https://doi.org/10.1007/s00464-016-4988-5 ).

Logic Programming (LP) is considered to be relatively simple for non-programmers, and allows the developer to focus on developing facts and rules of a logical derivation, and not on algorithms. Secure multiparty computation (MPC) is a methodology that allows several parties to process private data collaboratively without revealing the data to any party. In this paper, we bring together the notions of MPC and LP, allowing users to write privacy-preserving applications in logic programming language.

We propose new solutions that can efficiently address the problem of malicious memory access from pluggable computer peripherals and microcontrollers embedded within a system-on-chip. This problem represents a serious emerging threat to total-system computer security. Previous work has shown that existing defenses are insufficient and poorly deployed, in part due to performance concerns. In this paper we explore the threat and its implications for system architecture. We propose a range of protection techniques, from lightweight to heavyweight, across different classes of systems. We consider how emerging capability architectures (and specifically the CHERI protection model) can enhance protection and provide a convenient bridge to describe interactions among software and hardware components. Finally, we describe how new schemes may be more efficient than existing defenses.

Paper-based digital microfluidic biochip (PB-DMFB) technology provides a promising solution to many biochemical applications. However, PB-DMFB manufacturing process may suffer from potential security threats. For example, both Trojan insertion and man-in-the-middle attack may affect the functionality of PB-DMFBs. To ensure the correct functionality of PB-DMFBs, we propose a watermarking scheme to hides information in the PB-DMFB layout, which allows users to check design integrity and authenticate the source of the PB-DMFB design. As a result, it serves as a first countermeasure against both Trojan insertion and man-in-the-middle attacks for PB-DMFBs.

Video cameras are becoming ubiquitous in our daily lives. With the recent advancement of Artificial Intelligence (AI), live video analytics are enabling various useful services, including traffic monitoring and campus surveillance. However, current video analytics systems are highly limited in leveraging the enormous opportunities of the deployed cameras due to (i) centralized processing architecture (i.e., cameras are treated as dumb streaming-only sensors), (ii) hard-coded analytics capabilities from tightly coupled hardware and software, (iii) isolated and fragmented camera deployment from different service providers, and (iv) independent processing of camera streams without any collaboration. In this paper, we envision a full-fledged system for software-defined video analytics with cross-camera collaboration that overcomes the aforementioned limitations. We illustrate its detailed system architecture, carefully analyze the key system requirements with representative app scenarios, and derive potential research issues along with a summary of the status quo of existing works.

Storm and flood disasters such as typhoons and torrential rains are becoming more intense and frequent. The national government and municipalities must respond to such natural disasters as soon as possible. When the scale of damage is large; however, it takes much time to investigate the severity of damage, and the initial response can be delayed. If we could precisely and rapidly estimate the severity of damage for each city at an early stage, the national government would be able to better support the municipalities, and consequently respond quickly to help citizens. In this paper, we propose a novel approach to estimate the severity of disaster damage within a short time period after a disaster occurs by exploiting real-time population data generated from cellular networks. First, we investigate the relationship between real-time population data and the severity of damage. Then, we design a Graph Convolutional Networks for Disaster Damage Estimation, called D2E-GCN, which fully exploits the directed and weighted characteristics of human mobility graph. We conduct an offline evaluation on real-world datasets including two typhoons that hit Japan. The evaluation results show that the proposed method outperforms baseline methods which do not consider the graph structure of cities, and the proposed method can estimate the severity of damage approximately 48 hours after typhoons passed. Moreover, we find the experimental insight that the estimation performance can be significantly affected by the graph construction method for GCN models.

In federated learning (FL), model updates from mutually distrusting parties are aggregated in a centralized fusion server. The concentration of model updates simplifies FL's model building process, but might lead to unforeseeable information leakage. This problem has become acute due to recent FL attacks that can reconstruct large fractions of training data from ostensibly "sanitized" model updates. In this paper, we re-examine the current design of FL systems under the new security model of reconstruction attacks. To break down information concentration, we build TRUDA, a new cross-silo FL system, employing a trustworthy and decentralized aggregation architecture. Based on the unique computational properties of model-fusion algorithms, we disassemble all exchanged model updates at the parameter-granularity and re-stitch them to form random partitions designated for multiple hardware-protected aggregators. Thus, each aggregator only has a fragmentary and shuffled view of model updates and is oblivious to the model architecture. The deployed security mechanisms in TRUDA can effectively mitigate training data reconstruction attacks, while still preserving the accuracy of trained models and keeping performance overheads low.

Blind natural video quality assessment (BVQA), also known as no-reference video quality assessment, is a highly active research topic. In our recent contribution titled "Blind Natural Video Quality Prediction via Statistical Temporal Features and Deep Spatial Features" published in ACM Multimedia 2020, we proposed a two-level video quality model employing statistical temporal features and spatial features extracted by a deep convolutional neural network (CNN) for this purpose. At the time of publishing, the proposed model (CNN-TLVQM) achieved state-of-the-art results in BVQA. In this paper, we describe the process of reproducing the published results by using CNN-TLVQM on two publicly available natural video quality datasets.