Search

We investigate three tutorials on Android development to elicit design choices related to their organization and content. We analyze organization styles related to the intended reading sequence, use of code fragments, and reliance on external resources. We survey design variations in the choice of topics covered by tutorials, for example deprecated or convenience APIs, and concerns unrelated to the functional features such as development methodologies and best practices. We provide insights about the impact of different design decisions on readers. For example, a clear reading order provides an easier introduction to the technology for beginners, but focused, independent sections can be more effective for an intermediate-level audience. We describe these decision points and their trade-offs and propose guidelines to assist tutorial authors in making explicit and informed decisions about the design of tutorials.

In industry NLP application, our manually labeled data has a certain number of noisy data. We present a simple method to find the noisy data and re-label their labels to the result of model prediction. We select the noisy data whose human label is not contained in the top-K model’s predictions. The model is trained on the origin dataset. The experiment result shows that our method works. For industry deep learning application, our method improve the text classification accuracy from 80.5% to 90.6% in dev dataset, and improve the human-evaluation accuracy from 83.2% to 90.5%.

At present, the traditional geographical study is somewhat shifted to virtual web-datacentric geographical study and it was started almost around last decade of the twentieth century concerning remotely sensed earth surface, digitisation, virtual data storing and their artificial intelligence based machine learning for spatio-temporal analysis, prediction and thematic mapping. Now in the age of webdom—a global digital kingdom of internet and web services, when internet services flooded everything relating to continuous up-gradation of computer-related science and technologies along with increasing number of internet users and GPS enabled smart phones, geographical data are generated by exaflood leading to bigdata. Many online data are now easily available to everyone from anywhere at any time. The objective is to assess the webdom and to understand the scope of contemporary virtual web-data-centric geographical study involving geographical bigdata, machine learning and WebGIS, when virtual world abundantly expands gradually during this period of webdom. Each data in bigdata contains geospatial attributes. Worldwide under the exponential growth and advancement of data science like other scientific disciplines geography is also predicting different phenomenon based on artificial intelligence and machine learning. The three main facets of virtual web-data-centric geographical study are geographical bigdata, geographical machine learning, and WebGIS. It is centrally concerned with digital data and web services. Here data are analysed through machine learning methods. Sometimes reliability of web-data is in question without field verification which is considered as the primary requisition of traditional geography.

Machine learning teaches computers to think in a similar way to how humans do. An ML models work by exploring data and identifying patterns with minimal human intervention. A supervised ML model learns by mapping an input to an output based on labeled examples of input-output (X, y) pairs. Moreover, an unsupervised ML model works by discovering patterns and information that was previously undetected from unlabelled data. As an ML project is an extensively iterative process, there is always a need to change the ML code/model and datasets. However, when an ML model achieves 70-75% of accuracy, then the code or algorithm most probably works fine. Nevertheless, in many cases, e.g., medical or spam detection models, 75% accuracy is too low to deploy in production. A medical model used in susceptible tasks such as detecting certain diseases must have an accuracy label of 98-99%. Furthermore, that is a big challenge to achieve. In that scenario, we may have a good working model, so a model-centric approach may not help much achieve the desired accuracy threshold. However, improving the dataset will improve the overall performance of the model. Improving the dataset does not always require bringing more and more data into the dataset. Improving the quality of the data by establishing a reasonable baseline level of performance, labeler consistency, error analysis, and performance auditing will thoroughly improve the model's accuracy. This review paper focuses on the data-centric approach to improve the performance of a production machine learning model.

Peer evaluations are critical for assessing teams, but are susceptible to bias and other factors that undermine their reliability. At the same time, collaborative tools that teams commonly use to perform their work are increasingly capable of logging activity that can signal useful information about individual contributions and teamwork. To investigate current and potential uses for activity traces in peer evaluation tools, we interviewed (N=11) and surveyed (N=242) students and interviewed (N=10) instructors at a single university. We found that nearly all of the students surveyed considered specific contributions to the team outcomes when evaluating their teammates, but also reported relying on memory and subjective experiences to make the assessment. Instructors desired objective sources of data to address challenges with administering and interpreting peer evaluations, and have already begun incorporating activity traces from collaborative tools into their evaluations of teams. However, both students and instructors expressed concern about using activity traces due to the diverse ecosystem of tools and platforms used by teams and the limited view into the context of the contributions. Based on our findings, we contribute recommendations and a speculative design for a data-centric peer evaluation tool.

Digital supply chains (SCs) and data-centric collaborations have boosted data exchange between companies and, combined with recent advancements in data science, have brought a new type of knowledge...

Implementing reliable end-to-end transmission and reduced congestion in a single transport layer protocol is a critical issue for wireless sensor networks (WSNs). These issues degrade the performance of the WSN; since the devices used to configure the networks are resource-constrained ones. If the protocol manages without any one of the above-said problems which will not be a complete solution for transport layer issues in WSN. Some of the data-centric applications of WSNs need the reliable transmission of data packets with reduced congestion. To addresses the same, this paper proposed the data-centric transport layer protocol (DCTP). It effectively addresses problems like reliable delivery and congestion-free transmission. One of the major problems of WSNs is cluster head (CH) election and channel assignment to the members of CH. In this paper, the modified black widow optimization is used to select the CH which leads to reduced congestion and balanced energy utilization among the nodes. In a separate phase, the reliability of the event-to-sink transmission has also been addressed. Hence this proposed DCTP will be ultimately suitable for the data-centric applications of WSNs and ensures reliability and energy efficiency (i.e., 6.2315 mJ and 7.825 mJ) in the case of varying data rate and the number of nodes.

Body movements, from a short smile to a marathon run, are driven by muscle activity. Despite the fact that measuring muscle activity with electromyography (EMG) is technically well established, it is highly complex and its use in interfaces has been limited. Easy access to muscle sensing can offer new opportunities to Human-Computer Interaction (HCI) research. Off-the-shelf sensors often only provide low-level access, hence requiring expertise in signal processing and widening the gulf of execution for users without engineering skills. To address this challenge, we introduce EMBody, a data-centric toolkit for EMG-based interface prototyping and experimentation. EMBody offers multiple levels of prototyping fidelity for EMG sensing, signal processing, and data interpretation. Our data-centric toolkit encapsulates the different data representation stages, offering a wide range of customization opportunities to experts while also allowing non-technical designers to focus on creating new interaction techniques. EMBody features a lightweight form factor and wireless connectivity. Additionally, the system leverages an exploration-centered workflow by allowing rapid access to measurement data via the accompanying software. Users define a set of motions to be recognized and interactively provide example data points. The toolkit then handles signal processing and classification. The recognized movements are streamed on the local network, ready to be used by interactive applications. This paper reports on how to use EMBody and its implementation. We iteratively developed the toolkit in a series of workshops and example applications. Users who had none or very limited knowledge of EMG could rapidly create engaging functional prototypes, while experts appreciated the modularity of the software component allowing for a high degree of customization. We contribute the software and hardware components of EMBody as a tool for the research community to stimulate creative exploration of EMG systems.

The Internet of Things (IoT) boosts road safety, efficiency, and infotainment by connecting vehicles to form the Internet of Vehicles (IoV). Specifically to safety, IoV complements autonomous cars beyond sensors’ line-of-sight, facilitating vehicle-to-vehicle (V2V) communications in a smart transportation environment. The correctness of data exchanged among vehicles is paramount to ensure vehicles behave as per norms. Traditional misbehavior detection methods hardly defend vehicular security effectively due to rapid dynamics and location privacy. In particular, those node-centric classifiers become ill-fit in IoV. This work proposes a data-centric misbehavior detection model based on supervised machine learning (ML). The work also integrates plausibility checks with ML techniques and instantiates the model with six algorithms to demonstrate their comparative effectiveness. In addition to misbehavior detection, the model classifies attack types to support validating countermeasures. Specifically, the work analyzes the supervised learning algorithms for detecting misbehavior in IoV, compares their performance, and identifies their limitations. VeReMi, a vehicle-to-everything (V2X) position forgery attack built-in simulated road traffic data set, is used to test the effectiveness of the proposed model. The performance metrics include precision–recall (PR) and receiver operating characteristic (ROC) curves. The results demonstrate the effectiveness and significance of ML to detect misbehavior in IoV. The addition of plausibility checks improves the precision and recall by 5% and 2%, respectively.

In industry NLP application, our manually labeled data has a certain number of noisy data. We present a simple method to find the noisy data and relabel them manually, meanwhile we collect the correction information. Then we present novel method to incorporate the human correction information into deep learning model. Human know how to correct noisy data. So the correction information can be inject into deep learning model. We do the experiment on our own text classification dataset, which is manually labeled, because we relabel the noisy data in our dataset for our industry application. The experiment result shows that our method improve the classification accuracy from 91.7% to 92.5%. The 91.7% accuracy is trained on the corrected dataset, which improve the baseline from 83.3% to 91.7%.

Building an open global sensing layer is critical for the Internet of Things (IoT). In this article, we present a Sensory Data-Centric Networking (SDCN) architecture for inter-networking two main networked sensing systems in IoT—wireless sensor networks and mobile sensing networks. Specifically, the proposed SDCN is a systematic solution including NDNs for sensor nodes in the Zigbee network, NDNm for mobilephones in the Wi-Fi network, and NDNg for gateways. Considering the sensing requirement of IoT, we first design a novel Spatio-Temporal 16 Tree (ST16T) naming scheme associated with the scope-matching method. Based on the naming scheme, we further propose the related discovery methods, network switching mechanism, forwarding, and routing strategies according to the features of large-scale sensing and resource-constrained environment. A proof-of-concept prototype is implemented and further is deployed on our campus (BUPT) and the Great Wall (Shaanxi, China) for Environment Monitoring Project. Several experiments are conducted on the deployed platform. The experimental results show that SDCN outperforms the state-of-the-arts and gains a great performance improvement in terms of energy consumption, data collection efficiency, memory footprint, and time delay.

In this work, we describe an urban Internet of Things (IoT) architecture, grounded in big data patterns and focused on the needs of cities and their key stakeholders. First, the architecture of the dedicated platform USE4IoT (Urban Service Environment for the Internet of Things), which gathers and processes urban big data and extends the Lambda architecture, is proposed. We describe how the platform was used to make IoT an enabling technology for intelligent transport planning. Moreover, key data processing components vital to provide high-quality IoT data streams in a near-real-time manner are defined. Furthermore, tests showing how the IoT platform described in this study provides a low-latency analytical environment for smart cities are included.

In this present era, internet and cyber technologies have revolutionized and improved people's lifestyles completely at a remarkable rate. Having significant benefits, there are certain challenges faced by current technologies. In this perspective, one of the main challenges is not the production and acquisition of data; rather, it is the proper management, of it, including storage and analysis of data, has become a major concern. In this article, we conduct a study to discuss the two most widely used data centric technologies, i.e., Cyber Physical Systems (CPSs) and Internet of Things (IoT). The goal is to identify the architectural similarities and differences to achieve an interoperable solution that can manage data more efficiently. Our analysis revealed that a possible transformation of IoT to CPS and vice versa is possible when certain transformation steps are performed.

The efficiency of an accelerator depends on three factors-mapping, deep neural network (DNN) layers, and hardware-constructing extremely complicated design space of DNN accelerators. To demystify such complicated design space and guide the DNN accelerator design for better efficiency, we propose an analytical cost model, MAESTRO. MAESTRO receives DNN model description and hardware resources information as a list, and mapping described in a data-centric representation we propose as inputs. The data-centric representation consists of three directives that enable concise description of mappings in a compiler-friendly form. MAESTRO analyzes various forms of data reuse in an accelerator based on inputs quickly and generates more than 20 statistics including total latency, energy, throughput, etc., as outputs. MAESTRO's fast analysis enables various optimization tools for DNN accelerators such as hardware design exploration tool we present as an example.

In recent years, the number of financial technology players and users have increased at a significant rate due to the rapid technological advancement in financial technology. While smart devices are providing more useful features to users, they have also made it possible for cyber threats to migrate from desktops to smart devices. Thus, it is important for smart device users to be aware that their device could be exposed to cyber threats and that users could protect their devices by employing data-centric cyber security measures. This study reveals how financial technology business model responded to the breach phenomenon by employing data-centric protection approaches. The result is very interesting. Data-centric security is very needed as it is capable of protecting data as a whole. It provides a gapless protection, meaning to say, the data are encrypted and classified wherever it moves. With persistent protection and cross-platform operability, data-centric security will eliminate gaps and keep data protected.

The cyber-physical convergence, the fast expansion of the Internet at its edge, and tighter interactions between human users and their personal mobile devices push towards a data-centric Internet where the human user becomes more central than ever. We argue that this will profoundly impact primarily on the way data should be handled in the Next Generation Internet. It will require a radical change of the Internet data-management paradigm, from the current platform-centric to a human-centric model. In this paper we present a new paradigm for Internet data management that we name Internet of People (IoP) because it embeds human behavior models in its algorithms. To this end, IoP algorithms exploit quantitative models of the humans’ individual and social behavior, from sociology, anthropology, psychology, economics, physics. IoP is not a replacement of the current Internet networking infrastructure, but it exploits legacy Internet services as (reliable) primitives to achieve end-to-end connectivity on a global-scale. In this opinion paper, we first discuss the key features of the IoP paradigm along with the underlying research issues and challenges. Then, we present emerging data-management paradigms that are anticipating IoP.

Recent advances in machine learning, coupled with low-cost computation, availability of cheap streaming sensors, data storage and cloud technologies, has led to widespread multi-disciplinary research activity with significant interest and investment from commercial stakeholders. Mechanistic models, based on physical equations, and purely data-driven statistical approaches represent two ends of the modelling spectrum. New hybrid, data-centric engineering approaches, leveraging the best of both worlds and integrating both simulations and data, are emerging as a powerful tool with a transformative impact on the physical disciplines. We review the key research trends and application scenarios in the emerging field of integrating simulations, machine learning, and statistics. We highlight the opportunities that such an integrated vision can unlock and outline the key challenges holding back its realisation. We also discuss the bottlenecks in the translational aspects of the field and the long-term upskilling requirements of the existing workforce and future university graduates., 20 pages, 2 figures

The process of testing conventional programs is quite easy as compared to the programs using Deep Learning approach. The term Deep learning (DL) is used for a novel programming approach that is highly data centric and where the governing rules and logic are primarily dependent on the data used for training. Conventionally, Deep Learning models are evaluated by using a test dataset to evaluate their performance against set parameters. The difference in data and logic handling between programs using conventional methods and programs using the DL approach makes it difficult to apply the traditional approaches of testing directly to DL based programs. The accuracy of test data is currently the best measure of the adequacy of testing in the DL based systems. This poses a problem because of the difficulty in availability of test data that is of sufficient quality. This in turn restricts the level of confidence that can be established on the adequacy of testing of DL based systems. Unlike conventional applications, using the conventional programming approaches the lack of quality test data and the lack of interpretability makes the system analysis and detection of defects a difficult task in DL based systems. So testing of DL based models can be done automatically with a different approach compared to normal software.

The emergence of additive manufacture (AM) for metallic material enables components of near arbitrary complexity to be produced. This has potential to disrupt traditional engineering approaches. However, metallic AM components exhibit greater levels of variation in their geometric and mechanical properties compared to standard components, which is not yet well understood. This uncertainty poses a fundamental barrier to potential users of the material, since extensive post-manufacture testing is currently required to ensure safety standards are met. Taking an interdisciplinary approach that combines probabilistic mechanics and uncertainty quantification, we demonstrate that intrinsic variation in AM steel can be well described by a generative statistical model that enables the quality of a design to be predicted before manufacture. Specifically, the geometric variation in the material can be described by an anisotropic spatial random field with oscillatory covariance structure, and the mechanical behaviour by a stochastic anisotropic elasto-plastic material model. The fitted generative model is validated on a held-out experimental dataset and our results underscore the need to combine both statistical and physics-based modelling in the characterization of new AM steel products.

Pandey et al. (2021) demonstrate the importance of diversifying training data to make balanced predictions of thermodynamic properties for inorganic crystals.

Adapting applications’ data rates in multi-hop wireless ad-hoc networks is inherently challenging. Packet collision, channel contention, and queue buildup contribute to packet loss but are difficult to manage in conventional TCP/IP architecture. This work explores a data-centric approach based on Name Data Networking (NDN) architecture, which is considered more suitable for wireless ad-hoc networks. We show that the default NDN transport offers better performance in linear topologies but struggles in more extensive networks due to high collision and contention caused by excessive Interests from out-of-order data retrieval and redundant data transmission from improper Interest lifetime setting as well as in-network caching. To fix these, we use round-trip hop count to limit Interest rate and Dynamic Interest Lifetime to minimize the negative effect of improper Interest lifetime. Finally, we analyze the effect of in-network caching on transport performance and which scenarios may benefit or suffer from it.

In industry NLP application, our manually labeled data has a certain number of noisy data. We present a simple method to find the noisy data and relabel them manually, meanwhile we collect the correction information. Then we present novel method to incorporate the human correction information into deep learning model. Human know how to correct noisy data. So the correction information can be inject into deep learning model. We do the experiment on our own text classification dataset, which is manually labeled, because we relabel the noisy data in our dataset for our industry application. The experiment result shows that our method improve the classification accuracy from 91.7% to 92.5%. The 91.7% baseline is based on BERT training on the corrected dataset, which is hard to surpass.

The recent development of IOT emphasizes on the connection of most of the devices by means of smarter and intelligence way. These devices communicate with each other via internet without or/very less communication link of human being. The network of IoT is highly heterogeneous, uncontrolled, broad and perspicacious. All the data centric activities are done by 3 vital steps i.e sensing, analyzing and processing. The Internet of Things (IoT) is a cooperative system which connects a large number of computing devices which includes sensors, actuators, processors to a single IP address and the computing devices of IoT are provided with Unique Identifiers (UIDs). The major goal of IoT is to create a real world environment where frequently used agile objects coherently linked with internet. The growth rate of IoT devices has been increasing at a very high speed, so IOT can find its application in every field. All most all areas are using IoT to improve efficiency, customer service and value of the business, but in all respect IOT devices are still requires security and privacy optimization for future sustainable development.

Achieving the customer demand in manufacturing industries during Covid-19 situation is a challenging task since the industry needs to stick on to the local government policies such as high safety, work from home, man power reduction for the purpose of social distance, travelling in transport etc. If the machines produce quality products without any interruption, minor stoppages or breakdown, then the industry can achieve its production demand satisfying the customer need along with Covid-19 restrictions. By implementing autonomous maintenance called Jisu Hazen (JH) pillar of Total Productive Maintenance, around 70% of breakdown caused due to forced deterioration can be eliminated. Data centric digitalization through Industry 4.0 and Industrial Internet of Things, manpower and manmade errors are reduced which intern eliminate breakdown, defects and increase productivity. Using smart sensors, the signature parameters related to Clean Lubricate Inspection and Tightening CLIT are digitalized and a smart JH is followed. Through smart JH, operators working hours, number of employees are reduced, work from home situation is made for maintenance engineers to maintain Covid-19 rules.

The world is rapidly shifting towards data-centric computing model with significant growth in compute, storage and networking. Moore’s law scaling to reduce power, improve performance and reduce area and cost per transistor to support data centric growth places significant challenges for photomask process, control, and metrology. DUV lithography continues to be workhorse technology and continues to drive innovation. Multiple solutions are being developed to solve DUV mask challenges associated with a big die requirement. While DUV production is being projected to be strong for next few years, EUV is picking up pace and being deployed in high volume manufacturing. EUV is an opportunity-rich environment for improvements and innovations that will continue to advance the journey. EUV HiNA is a major step in the lithography paradigm where 0.55 NA EUV optics will enable a shift to sub-10nm resolution. The HiNA mask ecosystem will adapt by growing the newly created low NA EUV supply chain into new directions. The photomask environment will continue to bring creativity, drive innovations and gain commercial rewards.

There have been numerous calls for O&G companies to rethink the way they carry out capital projects. Various studies have highlighted that opportunities to enhance productivity exist from early development through to construction and handover to operations. Companies that are can realize these opportunities can complete projects safer and faster, reduce costs and quality issues, and improve schedule predictability. This paper presents data centric verification as a key enabler for such change, as it focusses on delivery of desired outcomes – the key to successful projects. This paper presents the rapid benefits that can be harvested, if the effort to put data centric verification in place is made. The following aspects are detailed to illustrate the benefits that are achievable: The wide variety of checking and confirmation activities on projects that frequently overlap, happen too late or don't focus on the correct aspects can be streamlined through a unified verification approach. Using a data centric approach to the set up and tracking of verification activities can support analytics to enable better planning and early identification of issues. Traceability in the data centric approach will enable quicker identification of root causes and remediations. Handover to operations moves from a document accumulation exercise to a data custody transfer. A typical O&G project is executed by a consortium of companies from owner/operators, engineering, procurement and construction companies, various suppliers and equipment manufacturers, etc. Data centric verification, if achieved, can only be through collaboration on a commonly agreed approach. Objectives/Scope Using data centric verification as an enabler to enhance project productivity from early development through handover. Methods, Procedures, Process This paper presents data centric verification as a key enabler for such change, as it focusses on delivery of desired outcomes – the key to successful projects. This paper presents the rapid benefits that can be harvested, if the effort to put data centric verification in place is made. The following aspects are detailed to illustrate the benefits that are achievable: The wide variety of checking and confirmation activities on projects that frequently overlap, happen too late or don't focus on the correct aspects can be streamlined through a unified verification approach. Using a data centric approach to the set up and tracking of verification activities can support analytics to enable better planning and early identification of issues. Traceability in the data centric approach will enable quicker identification of root causes and remediations. Handover to operations moves from a document accumulation exercise to a data custody transfer. Results, Observations, Conclusions The paper demonstrates the resulting requirements management and verification data models that can be established to enable the efficiency gains during project execution. Several example implementations of this data centric approach will be shared to validate the observed improvements. Novel/Additive Information Data centric verification is new to the Oil and Gas Industry. These Shell example projects are the first to implement this new approach within our supply chain.

Recent years have witnessed the rapid deployment of smart homes; most of them are controlled by remote servers in the cloud. Such designs raise security and privacy concerns for end users. In this paper, we describe the design of Sovereign, a home IoT system framework that provides end users complete control of their home IoT systems. Sovereign lets home IoT devices and applications communicate via application-named data and secures data directly. This enables direct, secure, one-to-one and one-to-many device-to-device communication over wireless broadcast media. Sovereign utilizes semantic names to construct usable security solutions. We implement Sovereign as a publish-subscribe-based development platform together with a prototype home IoT controller. Our preliminary evaluation shows that Sovereign provides a systematic, easy-to-use solution to user-controlled, self-contained smart homes running on existing IoT hardware without imposing noticeable overhead.

Data-Centric Networking (DCN) is a research project based on Named Data Networking (NDN), which focuses on the high-performance name-based forwarder, distributed pub/sub data distribution platform, distributed network storage, in-network processing platform, and blockchain-based data trading platform. In this paper, we present a design of real-time public Internet of Things (IoT) data distribution platform which is based on a Data-Centric Networking (DCN) distributed pub/sub data distribution platform.

Recent advances in sensor technology and wireless communications have led to many new data dissemination routing protocols, especially designed for wireless sensor networks, where energy awareness is the most important consideration. The focus of this thesis is on the area of routing protocols for wireless sensor networks, especially for those applications where, the whole sensor field need to be taken under observation to detect available different types of moving objects. Besides the efficient use of limited energy, reliability is another important issue in sensor communication, where the network is susceptible to environmental factors. In this thesis, the design of a new energy-efficient data-centric routing protocol, named Reliable and Energy-Efficient Protocol (REEP), is proposed. We have used MATLAB 7.4 for our implementation. The performance of REEP has been compared with Directed Diffusion (DD) for the aforementioned sensor network application. Our simulations and experimental results show that REEP performs better than DD.

Internet of Things (IoT) is bringing revolution into today’s world where devices in our surroundings become smart and perform daily-life activities and operations with more precision. The architecture of IoT is heterogeneous as it provides autonomy to nodes that they can communicate among other nodes and can also exchange information at any period. Due to the heterogeneous environment, IoT faces numerous security and privacy challenges, and one of the most significant challenges is the identification of malicious and compromised nodes. In this article, we have proposed a Machine Learning-based trust management approach for edge nodes. The proposed approach is a lightweight process to evaluate trust because edge nodes cannot perform complex computations. To evaluate trust, the proposed mechanism utilizes the knowledge and experience component of trust where knowledge is further based on several parameters. To eliminate the triumphant execution of good and bad-mouthing attacks, the proposed approach utilizes edge clouds, i.e., local data centers, to collect recommendations to evaluate indirect and aggregated trust. The trustworthiness of nodes is ranked between a certain limit where only those that satisfy the threshold value can participate in the network. To validate the performance of a proposed approach we have performed an extensive simulation in comparison with the existing approaches and the result shows the effectiveness of the proposed approach against several potential attacks.

Graphics Processing Units (GPUs) have become the default choice of acceleration in a wide range of application domains. To keep up with computational demands, the GPU memory system is constantly being innovated from both the cache and DRAM perspectives. Such innovations can adversely affect GPU reliability and in fact, can lead to an increase in the number of multi-bit faults. To address this problem, we systematically study a wide range of GPGPU applications and find that usually, only a small percentage of data needs protection to increase application resilience. This data is highly accessed and shared (constitutes hot memory), which implies that faults in this space can often lead to incorrect application output. An in-depth analysis of application code shows that information of such data can be passed on to the hardware to guide low-overhead detection/correction schemes. In this vein, we developed low-overhead partial data replication schemes that exploit latency tolerance in GPUs. Overall, this data-centric approach dramatically improves GPGPU application resilience, with a minimal additional average performance overhead of 1.2% for detection-only and 3.4% for detection-and-correction.

Identifying a suitable configuration of devices, software and infrastructures in the context of user requirements is fundamental to the success of delivering IoT applications. As possible configurations could be large in number and not all configurations are valid, a configuration knowledge representation model can provide ready-made configurations based on IoT requirements. Combining such a model within the context of a given user-oriented scenario, it is possible to automate the recommendation of solutions for deployment and long-time evolution of IoT applications. However, in the context of Cloud/Edge technologies, that may themselves exhibit significant configuration possibilities that are also dynamic in nature, a more unified approach is required. We present IoT-CANE (Context Aware recommendatioN systEm) as such a unified approach. IoT-CANE embodies a unified conceptual model capturing configuration, constraint and infrastructure features of Cloud/Edge together with IoT devices. The success of IoT-CANE is evaluated through an end-user case study.

Gateways play very important roles in smart homes. With gateways that can effectively separate applications and physical devices, the smart home architecture can be simplified. In the future, when more and more wireless devices are deployed in smart homes, gateways are facing critical challenges to accommodate the variety and heterogeneity of devices while maintaining enough simplicity, full interoperability, high energy and spectrum efficiency, and uncompromised system security. In this article, we first study the-state-of-the-art techniques and solutions for smart home infrastructure and present both their advantages and disadvantages. Then we propose a data-centric cognitive gateway, which is simple, efficient, and secure for smart homes. In addition, we investigate the trade-off among deployment cost, routine maintenance overhead, and system security. Our experiments verify the efficiency of our proposed solutions.

We aim to implement a Big Data/Extreme Computing (BDEC) capable system infrastructure as we head towards the era of Exascale computing - termed SAGE (Percipient StorAGe for Exascale Data Centric Computing). The SAGE system will be capable of storing and processing immense volumes of data at the Exascale regime, and provide the capability for Exascale class applications to use such a storage infrastructure. SAGE addresses the increasing overlaps between Big Data Analysis and HPC in an era of next-generation data centric computing that has developed due to the proliferation of massive data sources, such as large, dispersed scientific instruments and sensors, whose data needs to be processed, analysed and integrated into simulations to derive scientific and innovative insights. Indeed, Exascale I/O, as a problem that has not been sufficiently dealt with for simulation codes, is appropriately addressed by the SAGE platform. The objective of this paper is to discuss the software architecture of the SAGE system and look at early results we have obtained employing some of its key methodologies, as the system continues to evolve.

Modelling WSN data behaviour is relevant since it would allow to evaluate the capacity of an application for supplying the user needs, moreover, it could enable a transparent integration with different data-centric information systems. Therefore, this article proposes a data-centric UML profile for the design of wireless sensor nodes from the user point-of-view capable of representing the gathered and delivered data of the node. This profile considers different characteristics and configurations of frequency, aggregation, persistence and quality at the level of the wireless sensor nodes. Furthermore, this article validates the UML profile through a computer-aided software engineering (CASE) tool implementation and one case study, centred on the data collected by a real WSN implementation for precision agriculture and smart farming.

This paper proposes an innovative method of replicating the multi-modal public transport system in Singapore with high precision using smart card database. It replicates the operation of public transport system with known exogenous passenger demand and provides many operational details, including passenger inter-modal trip chains, operational timetable, and detailed transfer behaviour. The paper elaborates on the methodology of the replication including data cleaning, filtering, processing and converting the collected data to meaningful information such as bus journey trajectories and metro system timetable. Thereafter, actualised passenger trip chains are directly assigned to the replicated public transport supply. The resulting replication covers almost 96% of trips made in public transport in Singapore. It provides solid quantitative information on several aspects to support decision making, including precise temporal and spatial travel demand analysis, transfer pattern analysis, traffic condition investigation and bus utilisation analysis.

This paper investigates a data centric approach for future regulatory spectrum management (SM). Spectrum sensing data are collected by a spectrum environment awareness system built on a cloud-based service of Internet of Things. The data are used to characterize channel behaviors and establish a sharing predictor model which enables a set of efficient machine learning algorithms for automated spectrum sharing decision making. The performance of the decision process is evaluated, illustrating the feasibility and potential of this novel SM approach.

Virtualization has become a universal generalization layer in contemporary data centers. By multiplexing hardware resources into multiple virtual machines and facilitating several operating systems to run on the same physical platform at the same time, it can effectively decrease power consumption and building size or improve security by isolating virtual machines. In a virtualized system, memory resource supervision acts as a decisive task in achieving high resource employment and performance. Insufficient memory allocation to a virtual machine will degrade its performance drastically. On the contrasting, over allocation reasons ravage of memory resources. In the meantime, a virtual machine's memory stipulates may differ drastically. As a consequence, effective memory resource management calls for a dynamic memory balancer, which, preferably, can alter memory allocation in a timely mode for each virtual machine-based on their present memory stipulate and therefore realize the preeminent memory utilization and the best possible overall performance. Migrating operating system instances across discrete physical hosts is a helpful tool for administrators of data centers and clusters: It permits a clean separation among hardware and software, and make easy fault management. In order to approximate the memory, the stipulate of each virtual machine and to adjudicate probable memory resource disagreement, an extensively planned approach is to build an Least Recently Used based miss ratio curve which provides not only the current working set size but also the correlation between performance and the target memory allocation size. In this paper, the authors initially present a low overhead LRU-based memory demand tracking scheme, which includes three orthogonal optimizations: AVL based Least Recently Used association, dynamic hot set sizing. This assessment outcome confirms that, for the complete SPEC CPU 2006 benchmark set, subsequent to pertaining the 3 optimizing techniques, the mean overhead of MRC construction are lowered from 173% to only 2%. Based on current WSS, the authors then predict its trend in the near future and take different tactics for different forecast results. When there is an adequate amount of physical memory on the host, it locally balances its memory resource for the VMs. Once the local memory resource is insufficient and the memory pressure is predicted to sustain for a sufficiently long time, VM live migration, is used to move one or more VMs from the hot host to other host(s). Finally, for transient memory pressure, a remote cache is used to alleviate the temporary performance penalty. These experimental results show that this design achieves 49% center-wide speedup.

Part of the current smart cities research aims to efficiently measure its properties, improving our perceptions of objects, people, and the environment where we live. However, the implementation of sensing technologies in city scale is a critical factor in terms of cost and coverage. Thus, researchers have resorted to data-centric approaches to provide sensing systems based on big data. In this paper, we investigate data streams of online social networks to provide insights about people, venues, and city regions using a big data approach for urban sensing. We take advantage of the popularity of these applications to investigate social and spatiotemporal characteristics, demonstrating the feasibility of sensors based on data analytics.

Training datasets fundamentally impact the performance of machine learning (ML) systems. Any biases introduced during training (implicit or explicit) are often reflected in the system's behaviors leading to questions about fairness and loss of trust in the system. Yet, information on training data is rarely communicated to stakeholders. In this work, we explore the concept of data-centric explanations for ML systems that describe the training data to end-users. Through a formative study, we investigate the potential utility of such an approach, including the information about training data that participants find most compelling. In a second study, we investigate reactions to our explanations across four different system scenarios. Our results suggest that data-centric explanations have the potential to impact how users judge the trustworthiness of a system and to assist users in assessing fairness. We discuss the implications of our findings for designing explanations to support users’ perceptions of ML systems.

Many universities are introducing a new major in Data Science into their offering, to reflect the explosive growth in this field and the career opportunities it provides. As a field Data Science has elements from Computer Science and from Statistics, and curricula plans differ widely, both in the balance between the CS and Stats aspects, and also in the emphasis within the computing topics. This paper reports on the curriculum that has been taught for three years now at the University of Sydney. In particular, we describe the approach of a sequence of computing subjects which were developed specifically for the major, in order to bring students over several years to a sophisticated understanding of the data-handling aspects of Data Science. Students also take traditional subjects from both CS (such as Data Structures or AI) and from Statistics (such as Learning from Data and Statistical Inference). The data-centric specially-designed subjects we discuss in this paper are (i) Informatics: Data and Computation (in the first year), (ii) Big Data and Data Diversity (in the second year), and then upper-division subjects on (iii) Data Science Platforms, and (iv) Human-in-the-Loop Data Analytics.

The pharmaceutical industry is on the brink of entering into the digital age, yet still suffers from fundamental misconceptions and outdated IT systems that inhibit its progress. Four key criteria are identified that have enabled labs to reach the post-modern stage, which are insights generation through advanced analytics, automatic communication through machine to machine interfaces, removal of boundaries for an open lab, and novel means of ensuring trust through automatic submissions. Further progress in these four areas will enable the pharmaceutical laboratory to enter the digital age. Unfortunately, historical roadblocks in the form of an application-centric mindset have so far stifled progress. However, initiatives that supported other industries on their path into the digital age are introduced and evidences for the benefits of the digital age are provided. These span from advanced analytics, data-centric architecture, metadata supported communication, knowledge assisted submissions, to digital maturity models. It is concluded that executives and lab staff within Pharma needs a transition to a data-centric world view to reap all the benefits of the digital age for faster, better, and cheaper drug development.

The models and data framework demystifies the different roles that models and data play in software development and operation and clarifies where machine learning and artificial intelligence techniques could be used.

Serverless computing is a cloud-based execution paradigm that allows provisioning resources on-demand, freeing developers from infrastructure management and operational concerns. It typically involves deploying workloads as stateless functions that take no resources when not in use, and is meant to scale transparently. To make serverless effective, providers impose limits on a per-function level, such as maximum duration, fixed amount of memory, and no persistent local storage. These constraints make it challenging for data-intensive workloads to take advantage of serverless because they lead to sharing significant amounts of data through remote storage. In this paper, we build a performance model for serverless workloads that considers how data is shared between functions, including the amount of data and the underlying technology that is being used. The model's accuracy is assessed by running a real workload in a cluster using Knative, a state-of-the-art serverless environment, showing a relative error of 5.52%. With the proposed model, we evaluate the performance of data-intensive workloads in serverless, analyzing parallelism, scalability, resource requirements, and scheduling policies. We also explore possible solutions for the data-sharing problem, like using local memory and storage. Our results show that the performance of data-intensive workloads in serverless can be up to 4.32= faster depending on how these are deployed. This work was partially supported by the Ministry of Economy of Spain under contract TIN2015-65316-P, the Ministry of Science under contract PID2019-107255GB-C21/AEI/10.13039/501100011033, and the Generalitat de Catalunya under contract 2014SGR1051.

With the ever-growing usage of artificial intelligence in business decisions, organizations, and the day-to-day lives of the society at large, it is imperative that different forms of biases that human beings are privy to do not get trickled to the machine learning and artificial intelligence systems being developed and used in abundance.