Cross-domain Applications Research Articles

Charlotte: Reformulating Blockchains into a Web of Composable Attested Data Structures for Cross-Domain Applications

Cross-domain applications are rapidly adopting blockchain techniques for immutability, availability, integrity, and interoperability. However, for most applications, global consensus is unnecessary and may not even provide sufficient guarantees. We propose a new distributed data structure: Attested Data Structures (ADS), which generalize not only blockchains but also many other structures used by distributed applications. As in blockchains, data in ADSs is immutable and self-authenticating. ADSs go further by supporting application-defined proofs ( attestations ). Attestations enable applications to plug in their own mechanisms to ensure availability and integrity. We present Charlotte , a framework for composable ADSs. Charlotte deconstructs conventional blockchains into more primitive mechanisms. Charlotte can be used to construct blockchains but does not impose the usual global-ordering overhead. Charlotte offers a flexible foundation for interacting applications that define their own policies for availability and integrity. Unlike traditional distributed systems, Charlotte supports heterogeneous trust: different observers have their own beliefs about who might fail, and how. Nevertheless, each observer has a consistent, available view of data. Charlotte’s data structures are interoperable and composable : applications and data structures can operate fully independently or can share data when desired. Charlotte defines a language-independent format for data blocks and a network API for servers. To demonstrate Charlotte’s flexibility, we implement several integrity mechanisms, including consensus and proof of work. We explore the power of disentangling availability and integrity mechanisms in prototype applications. The results suggest that Charlotte can be used to build flexible, fast, composable applications with strong guarantees.

Discovering new applications: Cross-domain exploration of patent documents using causal extraction and similarity analysis

Determining a technology’s potential applications is crucial in assessing its level of innovation and evaluating its commercial viability. However, patent documents offer limited insights into a technology’s full potential. As a solution, this study suggests an approach to explore a technology’s applicability beyond what is explicitly stated in patents. The approach employs causal extraction to extract sentences expressing technologies and their applications from patents, followed by deep learning-based similarity analysis to compare the similarity of these sentences. Experimental results show its effectiveness in extracting sentences about technologies and applications and its superiority in terms of F1 score compared to benchmark models. This study enables cross-domain comparisons of technologies and applications, identifies multiple prospective applications for a given technology, and offers new opportunities for patent value analysis and intellectual property management in the industry. A cross-domain application network of the proposed method demonstrates how to find all potential cross-domain connections of a given data and we provide open access to the code.

BMP: An Evaluation System for Bitcoin Mining Pools

With the in-depth research and cross-domain applications of blockchain in various industries, blockchain mining attracts great attention. A comprehensive and in-depth evaluation of mining pool is essential to the Bitcoin network supervision as well as the reduction of carbon emission. However, due to the difficulty of obtaining mining pool internal data, the current evaluation is limited to a single mining power indicator. In this article, we build a four-layer mining pool evaluation model, named BMP, based on the analytic hierarchy process (AHP). The model contains three first-level indicators and nine second-level indicators, covering various aspects of the performance of mining pool. Miner address extraction provides basic data for the calculation of these indicators. By utilizing the model proposed, we evaluated and ranked eight mainstream mining pools, which broadcasted 71.69% of the blocks during the period January 1, 2021-July 31, 2021. Through comparative analysis of evaluation result, we find that the 65% decline of mining power in June did not pose a threat to the stability of Bitcoin system. In addition, we also revealed the reasons for various fluctuations.

A Cross-Domain Weakly Supervised Diabetic Retinopathy Lesion Identification Method Based on Multiple Instance Learning and Domain Adaptation.

Accurate identification of lesions and their use across different medical institutions are the foundation and key to the clinical application of automatic diabetic retinopathy (DR) detection. Existing detection or segmentation methods can achieve acceptable results in DR lesion identification, but they strongly rely on a large number of fine-grained annotations that are not easily accessible and suffer severe performance degradation in the cross-domain application. In this paper, we propose a cross-domain weakly supervised DR lesion identification method using only easily accessible coarse-grained lesion attribute labels. We first propose the novel lesion-patch multiple instance learning method (LpMIL), which leverages the lesion attribute label for patch-level supervision to complete weakly supervised lesion identification. Then, we design a semantic constraint adaptation method (LpSCA) that improves the lesion identification performance of our model in different domains with semantic constraint loss. Finally, we perform secondary annotation on the open-source dataset EyePACS, to obtain the largest fine-grained annotated dataset EyePACS-pixel, and validate the performance of our model on it. Extensive experimental results on the public dataset FGADR and our EyePACS-pixel demonstrate that compared with the existing detection and segmentation methods, the proposed method can identify lesions accurately and comprehensively, and obtain competitive results using only coarse-grained annotations.

A data-driven framework for building energy benchmarking and renovation decision-making support in Sweden

In Europe, the buildings sector is responsible for 40% of energy use and more than 30% of buildings are older than 50 years. Due to ageing, a large number of houses require energy-efficient renovation to meet building energy performance standards and the national energy efficiency target. Although Swedish house owners are willing to improve energy efficiency, there is a need for a dedicated platform providing decision-making knowledge for house owners to benchmark their buildings. This paper proposes a data-driven framework for building energy renovation benchmarking as part of an energy advisory service development for the Västerbotten region, Sweden. This benchmark model facilitates regional homeowners to benchmark their building energy performance relative to the national target and similar neighbourhood buildings. Specifically, based on user input data such as energy use, location, construction year, floor area, etc., this model benchmarks the user’s building performance using two benchmark references i.e., 1) Sweden’s target to reduce buildings by 50% energy use intensity (EUI) by 50% by 2050 compared to the average EUI in 1995, 2) comparing user building with the most relevant peer group of buildings, using energy performance certificates (EPC) big data. Several building groups will be classified based on influential factors that affect building energy use. Hence, this benchmark provides decision-making supportive knowledge to homeowners e.g., whether they need to perform an energy-efficient renovation. In the future, this methodology will be extended and implemented in the digital platform to provide helpful insights to decide on suitable EEMs. This work is an integral part of project AURORAL aims to deliver an interoperable, open, and integrated digital platform, demonstrated by cross-domain applications through large-scale pilots in 8 regions in Europe, including Västerbotten.

Peridinin-chlorophyll-protein complex industry from algae: A critical review of the current advancements, hurdles, and biotechnological potential

PCPs (Peridinin-chlorophyll a-proteins), water-soluble light-harvesting complexes only found in dinoflagellates, have achieved tremendous research progress in the past few decades. Their commercial form product, Peridinin-chlorophyll-protein complex, is enormously applied in fluorescent label probes because of unique fluorescence characteristic. The huge application prospect has spawned a large number of studies on PCPs. Most of the research has been carried out around the light factor, however, other factors and the regulatory mechanism of PCPs are still indistinct. Although the complete nucleic acid sequences, vivid protein structures and explicit energy transfer mechanism of some PCPs have been revealed, that of PCPs in other dinoflagellates are not yet clear, which are the trends in future. It is not difficult to realize the crude purification of PCPs, but pioneering new ways and obtaining high-purity PCPs are still difficulties. Many applications of PCPs have been developed, nevertheless, plenty of which are in the laboratory stage. Due to the variety of research results of PCPs but lack of systematic summary, we carefully review the recent information about PCPs in fields of PCP-producing algae species, metabolic regulation, structure, energy transfer and photoprotection. Moreover, we further highlight purification processes and potential applications of PCPs. Finally, the challenges and strategies of large-scale commercial and cross-domain applications of PCPs are also analyzed and assessed.

Learning data science in elementary school mathematics: a comparative curriculum analysis

BackgroundData literacy is increasingly important in today’s data-driven world. Students across many educational systems first formally learn about data in elementary school not as a separate subject but via the mathematics curriculum. This experience can create tensions in the priorities of learning and assessment given the presence of other foundational mathematics content domains such as numbers, algebra, measurement, and geometry. There is a need to study data literacy in comparison to these other content domains in elementary mathematics. To address this need, we developed a methodology motivated by thinking curriculum theory and aligned with international assessment framework, for comparative analysis across mathematics content domains. This methodology examined increasing levels of cognitive domains from knowing to applying to reasoning across mathematics content domains. Intended, assessed, and attained curricula were analyzed using Singapore as a case study, combined with broader comparisons to attainments in four East Asian countries in TIMSS, an international large-scale assessment.ResultsWe found that learning in the data domain had very limited coverage in intended and assessed curricula in Singapore. However, compared to other mathematics content domains, the data curriculum placed heavier emphasis on higher-order cognitive domains including the use of generally difficult mixed data visualizations. This demanding curriculum in Singapore was associated with the highest attainment in the data domain among average 4th grade Singaporean students relative to students in four East Asian countries in TIMSS, as analyzed by quantile regression. However, lower-performing Singaporean students at the 10th percentile generally did not outperform their East Asian peers. We further found very limited applications of data in other mathematics domains or cross-domain learning more generally.ConclusionOur study offers a comparative analysis of the data curriculum in elementary school mathematics education. While the data curriculum was cognitively demanding and translated to very high average attainments of Singaporean students, the curriculum did not equally help weaker Singaporean students, with implications on current discourse on equity–excellence trade-off in science, technology, engineering, and mathematics (STEM) education. Our study further highlights the lack of cross-domain learning in mathematics involving data. Despite the broad applicability of data science, elementary school students’ first formal experience with data may lack emphasis on its cross-domain applications, suggesting a need to further integrate data skills and competencies into the mathematics curriculum and beyond.

Training a Machine Learning Model for representing Manufacturing Systems towards optimizing Resilience

Studies affecting manufacturing system resilience have been of rising interest in recent years due to increased rates of naturally caused, and socio-economic disturbances. The non-linearity in modern manufacturing systems, attributed to a host of internal and external complexities in structure and operations cannot be properly modeled using conventional white-box techniques. This paper focuses on building a black-box description of a manufacturing system as a solution, addressing the inherent non-linearity, and further supporting fast investigations on a larger solution space with the example of an application scenario. This work acts as a precursor towards the implementation of a conceptual workflow describing a cross-domain application of machine learning (ML) for optimizing manufacturing system resilience.

Deep Learning for Clothing Style Recognition Using YOLOv5.

With the rapid development of artificial intelligence, much more attention has been paid to deep learning. However, as the complexity of learning algorithms increases, the needs of computation power of hardware facilities become more crucial. Instead of the focus being on computing devices like GPU computers, a lightweight learning algorithm could be the answer for this problem. Cross-domain applications of deep learning have attracted great interest amongst researchers in academia and industries. For beginners who do not have enough support with software and hardware, an open-source development environment is very helpful. In this paper, a relatively lightweight algorithm YOLOv5s is addressed, and the Google Colab is used for model training and testing. Based on the developed environment, many state-of-art learning algorithms can be studied for performance comparisons. To highlight the benefits of one-stage object detection algorithms, the recognition of clothing styles is investigated. The image samples are selected from datasets of fashion clothes and the web crawling of online stores. The image data are categorized into five groups: plaid; plain; block; horizontal; and vertical. Average precison, mean average precison, recall, F1-score, model size, and frame per second are the metrics used for performance validations. From the experimental outcomes, it shows that YOLOv5s is better than other learning algorithms in the recognition accuracy and detection speed.

Finite Aperture Stereo

Multi-view stereo remains a popular choice when recovering 3D geometry, despite performance varying dramatically according to the scene content. Moreover, typical pinhole camera assumptions fail in the presence of shallow depth of field inherent to macro-scale scenes; limiting application to larger scenes with diffuse reflectance. However, the presence of defocus blur can itself be considered a useful reconstruction cue, particularly in the presence of view-dependent materials. With this in mind, we explore the complimentary nature of stereo and defocus cues in the context of multi-view 3D reconstruction; and propose a complete pipeline for scene modelling from a finite aperature camera that encompasses image formation, camera calibration and reconstruction stages. As part of our evaluation, an ablation study reveals how each cue contributes to the higher performance observed over a range of complex materials and geometries. Though of lesser concern with large apertures, the effects of image noise are also considered. By introducing pre-trained deep feature extraction into our cost function, we show a step improvement over per-pixel comparisons; as well as verify the cross-domain applicability of networks using largely in-focus training data applied to defocused images. Finally, we compare to a number of modern multi-view stereo methods, and demonstrate how the use of both cues leads to a significant increase in performance across several synthetic and real datasets.

Yin-Yang: Programming Abstractions for Cross-Domain Multi-Acceleration.

FPGA accelerators offer performance and efficiency gains by narrowing the scope of acceleration to one algorithmic domain. However, real-life applications are often not limited to a single domain, which naturally makes Cross-Domain Multi-Acceleration a crucial next step. The challenge is, existing FPGA accelerators are built upon their specific vertically-specialized stacks, which prevents utilizing multiple accelerators from different domains. To that end, we propose a pair of dual abstractions, called Yin-Yang, which work in tandem and enable programmers to develop cross-domain applications using multiple accelerators on a FPGA. The Yin abstraction enables cross-domain algorithmic specification, while the Yang abstraction captures the accelerator capabilities. We also develop a dataflow virtual machine, dubbed XLVM, that transparently maps domain functions (Yin) to best-fit accelerator capabilities (Yang). With six real-world cross-domain applications, our evaluations show that Yin-Yang unlocks 29.4× speedup, while the best single-domain acceleration achieves 12.0×.

A comprehensive review on bio-inspired flower pollination algorithm

Since time immemorial, the complex and highly nonlinear nature of various optimization problems in science and engineering has motivated researchers to develop sophisticated solutions. Along with traditional algorithms, in recent years, many computationally intelligent meta-heuristic techniques have also grabbed attention due to their effectiveness and flexibility. Flower Pollination Algorithm (FPA) is one such biologically-motivated procedure, which is influenced by the fertilization procedure of flowering species of plants. It has been effectively incorporated to fix many real-world problems in fields like wireless sensor networking, signal processing, power and communications, rain fall prediction, global function optimization and many more. Keeping in mind, its growing popularity due to its simplicity, adaptability and cross domain applicability, in this paper, a comprehensive review on FPA is presented. It throws light on its biological inspiration, algorithm, variants and applications in various industries.

Federated Transfer Learning Based Cross-Domain Prediction for Smart Manufacturing

Smart manufacturing aims to support highly customizable production processes. Therefore, the associated machine intelligence needs to be quickly adaptable to new products, processes, and applications with limited training data while preserving data privacy. In this article, a new federated transfer learning framework, federated transfer learning for cross-domain prediction, is proposed to address the challenges of data scarcity and data privacy faced by most machine learning approaches in modern smart manufacturing with cross-domain applications. The framework architecture consists of a central server and several groups of smart devices, where each group handles a different application. The existing applications can share their knowledge through the central server as base models, while new applications can convert a base model to their target-domain models with limited application-specific data using a transfer learning technique. Meanwhile, the federated learning scheme is deployed within a group to further enhance the accuracy of the application-specific model. The integrated framework allows model sharing across the central server and different smart devices without exposing any raw data and, hence, protects the data privacy. Two public datasets, COCO and PETS2009, which represent the source and target applications, are employed for evaluations. The simulation results show that the proposed method outperforms two state-of-the-art machine learning approaches by achieving better learning efficiency and accuracy.

ML-WiGR: a meta-learning-based approach for cross-domain device-free gesture recognition

Accurate sensing and understanding of gestures can improve the quality of human–computer interaction and show great theoretical significance and application potentials in the fields of smart home, assisted medical care and virtual reality. WiFi channel state information (CSI)-based device-free wireless gesture recognition requires no sensors and has a series of advantages such as permission for non-line-of-sight scenario, low cost, preserving for personal privacy and working in the dark night. Although most of the current WiFi CSI-based gesture recognition approaches can achieve good performance, they are difficult to adapt to the new domains. Therefore, this paper proposes ML-WiGR, a novel approach for device-free gesture recognition in cross-domain applications. ML-WiGR applies convolutional neural networks (CNN) and long short-term memory (LSTM) neural networks as the basic model for gesture recognition to extract spatial and temporal features. Combined with the meta-learning training mechanism, ML-WiGR can dynamically adjust the learning rate and meta-learning rate in training process adaptively and optimize the initial parameters of a basic model for gesture recognition, only using a few samples and several iterations to adapt to the new domain. In the experiments, the approach is tested under a variety of scenarios. The results show that ML-WiGR can achieve comparable performance against existing approaches with only a small number of samples for training in cross-domains.

Performance enhancement of futuristic airplanes by nature inspired biomimetic fish scale arrays—A design approach

Biomimetics has an immense potential to drive the next generation of technologies forward by propounding competent solutions from nature. For decades, the nonsmooth topography of most living creatures has been critical to their existence and survival. Contrary to human-made smooth surfaces, when adapted for fluid flow applications, nonsmooth surfaces can enhance the overall aerodynamic performance by reducing the drag force. Recently, the bioinspired scale structure of fish has been identified as a key biomimetic derivative for improving the aerodynamic efficiency in various cross-domain applications. This study investigates the aerodynamics of a fish scale array (FSA) NACA 0021 model at a specific Reynolds number (Re) of approximately 2.46 × 105 that is meant for laminar flow conditions using Computational Fluid Dynamics (CFD) tools and a subsonic wind tunnel facility. A 3D printed biomimetic FSA film is developed and affixed on the NACA 0021 wing profile for the experimental investigation, followed by flow visualization through the smoke tunnel facility. As proved qualitatively under specific aerodynamic conditions, the creation of velocity streaks has a major role in the drag reduction process. To obtain a clear perspective of flow across the overlapping fish scale structures, the results are focused on the central and overlapping regions of the FSA structure. The experiment has proved the existence of a combined formation of low and high velocity streaks in central and overlapping regions. The FSA 0021 model showed a maximum drag reduction of 9.57%, which was attributed to the streak formation phenomenon observed in the overlapping FSA configuration.

A survey on extraction of causal relations from natural language text

As an essential component of human cognition, cause–effect relations appear frequently in text, and curating cause–effect relations from text helps in building causal networks for predictive tasks. Existing causality extraction techniques include knowledge-based, statistical machine learning (ML)-based, and deep learning-based approaches. Each method has its advantages and weaknesses. For example, knowledge-based methods are understandable but require extensive manual domain knowledge and have poor cross-domain applicability. Statistical machine learning methods are more automated because of natural language processing (NLP) toolkits. However, feature engineering is labor-intensive, and toolkits may lead to error propagation. In the past few years, deep learning techniques attract substantial attention from NLP researchers because of its powerful representation learning ability and the rapid increase in computational resources. Their limitations include high computational costs and a lack of adequate annotated training data. In this paper, we conduct a comprehensive survey of causality extraction. We initially introduce primary forms existing in the causality extraction: explicit intra-sentential causality, implicit causality, and inter-sentential causality. Next, we list benchmark datasets and modeling assessment methods for causal relation extraction. Then, we present a structured overview of the three techniques with their representative systems. Lastly, we highlight existing open challenges with their potential directions.

An Ontology – based Contextual Approach for Cross-domain Applications in Internet of Things

The Internet of Things is an ecosystem which enables objects and devices such as sensors or actuators to communicate and exchange information with each other without human intervention. One of the main challenges in the Internet of Things is the lack of semantic interoperability; devices cannot understand the meaning of raw data, due to the diversity and heterogeneity in data formats from different sources. In order to deal with semantic interoperability, the ontologies are the one way to integrate semantics to raw data; they describe an IoT system and represent the data in a standardized way. The IoT devices provide a great deal of IoT data, mainly used for specific IoT applications such as smart home, smart farming, smart cities or healthcare. Therefore, existing applications became isolated in vertical silos, each one of them use independently their own model (i.e. ontology), which makes this ontologies also limited to a specific domain. Our approach has the goal of breaking down these vertical silos and achieves a semantic interoperability across IoT domains in cross-domain applications. In this paper, we have proposed a development of a single cross-domain ontology named CDOnto, it is considered to be a generic across different IoT domains, which can be extended by domain-specific ontologies. The proposed model follows a contextual approach to organize and distinguish the combined domains (i.e. contexts) representations. In addition, the ontology allows reasoning across overlapping domains and infers a complementary and new knowledge required in cross-domain applications.

How can generative adversarial networks impact computer generated art? Insights from poetry to melody conversion

Recent advances in deep learning and generative adversarial networks (GANs), in particular, has enabled interesting applications including photorealistic image generation, image translation, and automatic caption generation. This has opened up possibilities for many cross-domain applications in computer generated arts and literature. Although there are existing software-based approaches for generating musical accompaniment of a given poetry, there are no existing implementation using GANs. This work proposes a novel poetry to melody generation conditioned on poem emotion using GANs. A dataset containing pairs of poetry and melody based on three emotion categories is introduced. Furthermore, various GAN architectures including SpecGAN and WaveGAN were explored for automatic melody synthesis for a given class of poetry. Conditional SpecGAN produced the best melodies according to quantitative metrics. Melodies produced by SpecGAN were evaluated by volunteers who deemed the quality to be above average.

A Sustainable Method for Publishing Interoperable Open Data on the Web

Smart cities need (sensor) data for better decision-making. However, while there are vast amounts of data available about and from cities, an intermediary is needed that connects and interprets (sensor) data on a Web-scale. Today, governments in Europe are struggling to publish open data in a sustainable, predictable and cost-effective way. Our research question considers what methods for publishing Linked Open Data time series, in particular air quality data, are suitable in a sustainable and cost-effective way. Furthermore, we demonstrate the cross-domain applicability of our data publishing approach through a different use case on railway infrastructure—Linked Open Data. Based on scenarios co-created with various governmental stakeholders, we researched methods to promote data interoperability, scalability and flexibility. The results show that applying a Linked Data Fragments-based approach on public endpoints for air quality and railway infrastructure data, lowers the cost of publishing and increases availability due to better Web caching strategies.

Gigahertz Low-Loss and High Power Handling Acoustic Delay Lines Using Thin-Film Lithium-Niobate-on-Sapphire

In this work, we present the first group of gigahertz low-loss, wideband, and high power handling delay lines (ADLs) using a thin-film lithium niobate (LiNbO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sub> )-on-sapphire platform. The ADLs leverage a single-phase unidirectional transducer (SPUDT) to efficiently excite the shear horizontal surface acoustic wave (SH-SAW) in the film stack. The fabricated miniature SH-SAW ADL at 1.1 GHz shows a low insertion loss (IL) of 2.8 dB, a wide fractional bandwidth (FBW) of 6.14%, and a fast phase velocity of 5127 m/s. The device also features a high 1-dB compression point (P1dB) of 30.4 dBm. The temperature coefficient of frequency is -45 ppm/K. ADLs with delays between 12 and 172 ns have been implemented, with IL between 2.8 and 8.3 dB. SH-SAW propagation characteristics are extracted, showing a group velocity of 4747 m/s and a propagation loss of 6.73 dB/mm or 31.9 dB/μs. The simultaneous low-loss and high power handling illustrate the great potential of LiNbO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sub> -on-sapphire for RF and cross domain applications at gigahertz.