Search

Sewer networks play a pivotal role in our everyday lives by transporting the stormwater and urban sewage away from the urban areas. In this regard, Sewer Overflow (SO) has been considered as a detrimental threat to our environment and health, which results from the wastewater discharge into the environment. In order to grapple with such deleterious phenomenon, numerous studies have been conducted; however, there has not been any review paper that provides the researchers undertaking research in this area with the following inclusive picture: (1) detailed-scientometric analysis of the research undertaken hitherto, (2) the types of methodologies used in the previous studies, (3) the aspects of environment impacted by the SO occurrence, and (4) the gaps existing in the relative literature together with the potential future works to be undertaken. Based on the comprehensive review undertaken, it is observed that simulation and artificial intelligence-based methods have been the most popular approaches. In addition, it has come to the attention that the detrimental impacts associated with the SO are fourfold as follows: air, quality of water, soil, and business and structure. Among these, the majority of the studies' focus have been tilted towards the impact of SO on the quality of ground water. The outcomes of this state-of-the-art review provides the researchers and environmental engineers with inclusive hindsight in dealing with such serious issue, which in turn, this culminates in a significant improvement in our environment as well as humans' well-beings. • A detailed scientometric analysis of the research carried out on sewer overflow hitherto is undertaken. • All different types of methodologies used in the previous studies are thoroughly investigated. • The aspects of environment impacted by the SO occurrence are reviewed. • The gaps existing in the relative literature together with the potential future works to be undertaken are put forward. [ABSTRACT FROM AUTHOR]

• Application of ML methods in lasers is limited by a number of significant factors. • ML methods is often claimed where simple automation of control is introduced. • It is not clear, what kind of advantages will result from ML technologies in lasers. • Penetration of ML methods is frequently limited by the relatively low level of AI. • Presently, ML is not a universal method of photonics. Machine learning methods are being successfully applied in various domains of human activity, including optics. Can they become a universal instrument capable of improving user aspect of photonic devices? Is their application a necessary element of such improvement? Answers to fundamental questions arising from application of machine learning methods in ultrashort-pulsed lasers are discussed. Future prospects and current fundamental limitations of such methods are analysed. [ABSTRACT FROM AUTHOR]

Sir Ronald Aylmer Fisher opened many new areas in Multivariate Analysis, and the one which we will consider is discriminant analysis. Several papers by Fisher and others followed from his seminal paper in 1936 where he coined the name discrimination function. Historically, his four papers on discriminant analysis during 1936–1940 connect to the contemporaneous pioneering work of Hotelling and Mahalanobis. We revisit the famous iris data which Fisher used in his 1936 paper and in particular, test the hypothesis of multivariate normality for the data which he assumed. Fisher constructed his genetic discriminant motivated by this application and we provide a deeper insight into this construction; however, this construction has not been well understood as far as we know. We also indicate how the subject has developed along with the computer revolution, noting newer methods to carry out discriminant analysis, such as kernel classifiers, classification trees, support vector machines, neural networks, and deep learning. Overall, with computational power, the whole subject of Multivariate Analysis has changed its emphasis but the impact of this Fisher's pioneering work continues as an integral part of supervised learning in Artificial Intelligence (AI). [ABSTRACT FROM AUTHOR]

Adequate protection of the environment is one of the hot spots of concern for all sectors of society due to severe environmental pollution. The solution to this issue is friendly management of the environment. With the rapid growth of Chinese Manufacturing SMEs for economic development, environmental pollution and abuse of resources are arising. To resolve these issues, Chinese manufacturing SMEs are accelerating the implementation of green innovation in their industries. However, it is a complex task that involves enterprise, government, and social considerations. Therefore, it is essential to identify the green drivers for this implementation. With a focus on China's current situation from previous research and views from experts, this study aims to investigate how Chinese Manufacturing Small and Medium-sized Enterprises (SMEs) are responding to resource misuse and environmental pollution by implementing green innovation, emphasising the role of artificial intelligence (AI) in improving environmental performance. This study primarily looks into the factors that influence the adoption of green innovations by analysing the growth paths of Chinese SMEs operating in highly polluting industries over a longer time frame than five years. Artificial Intelligence is a valuable tool for solving the issues of ecological degradation. A quantitative method has been implemented for the Chinese companies' samples from the deeply polluting industries for more than five years. The findings of this paper advise that the average board size, the governing board meetings, and organizational performance are positively connected with the Chinese firms' environmental process. Board independence and diversity of gender have irrelevant associations with ecological performance. A convenient threshold regression model has been used to accumulate the respondents' data. It also reveals that larger board sizes and more frequent governing board meetings are positively associated with improved environmental performance among these firms. The findings state the critical implications for the firm executives, policymakers, environmental activists, and regulators. This result supports the insight drained from the resource dependence, stakeholder, firm agency, and legitimacy theories. • The paper tackles severe environmental pollution and the need for eco-friendly management practices. • The paper emphasizes the need for these enterprises to adopt green innovation practices. • Implementing green innovation is a complex task requiring coordination between enterprises, government, and society. • The paper highlights AI's role as a valuable tool in addressing ecological degradation issues. • The research utilized a threshold regression model to analyse respondent data effectively. [ABSTRACT FROM AUTHOR]

Deep learning models are being used for natural language processing. Despite their success, these models have been employed for only a few languages. Pretrained models also exist but they are mostly available for the English language. Low resource languages like Urdu are not able to benefit from these pre-trained deep learning models and their effectiveness in Urdu language processing remains a question. This paper investigates the usefulness of deep learning models for the next word prediction and suggestion model for Urdu. For this purpose, this study considers and proposes two word prediction models for Urdu. Firstly, we propose to use LSTM for neural language modeling of Urdu. LSTMs are a popular approach for language modeling due to their ability to process sequential data. Secondly, we employ BERT which was specifically designed for natural language modeling. We train BERT from scratch using an Urdu corpus consisting of 1.1 million sentences thus paving the way for further studies in the Urdu language. We achieved an accuracy of 52.4% with LSTM and 73.7% with BERT. Our proposed BERT model outperformed two other pre-trained BERT models developed for Urdu. Since this is a multi-class problem and the number of classes is equal to the vocabulary size, this accuracy is still promising. Based on the present performance, BERT seems to be effective for the Urdu language, and this paper lays the groundwork for future studies. • The first language model for Urdu using LSTM and BERT. • The proposed model predicts the next word in Urdu language. • Presents a BERT for Urdu language that is trained from scratch on 1.1 million Urdu sentence. • Present a pre-trained language model for use in many NLP tasks for Urdu. • The BERT model trained from scratch outperforms other pre-trained BERT models developed for Urdu. [ABSTRACT FROM AUTHOR]

• Minimization of both cloud service cost and request peak is addressed using game theory. • Nash equilibrium of the formalized game is proven theoretically. • A distributed Peak Clipping Algorithm (DPCA) is proposed to calculate the Nash equilibrium solution. • Experiments validate the request peak and the payment cost of customer are reduced simultaneously, thereby increasing the benefit. The rapid development of Internet of Things (IoT) technology and the popularity of Artificial Intelligence (Al) technology research have brought new opportunities for the development of cloud computing (CC). With the increasing number of mobile Internet access devices and IoT access devices, the number of task requests from CC customers for AI services in the network has also experienced an explosive growth. In this paper, the focus is on the possible overload of cloud providers during the peak period of cloud service requests. Thetime attributes of cloud task execution are classified to avoid overloading the cloud provider as much as possible. In a distributed cloud environment, it is necessary to consider the time flexible attributes of cloud tasks to reasonably compete for cloud resources. In this work, game theory (GT) is introduced to formulate a cloud service scheduling game, in which participants are cloud customers who participate in the purchase of cloud services. The players' strategies are the time flexibility of each cloud task. The problem is formulated as minimizing the cost of scheduling cloud services and a noncooperative game among the customers (as players) is presented. Then the existence of the Nash equilibrium (NE) solution of the game has been proved and a new algorithm has been proposed in this paper to compute it. In addition, the analysis process of the convergence of the proposed PCA algorithm and the proof of its convergence to NE are also included in this paper. At the end of the paper, simulations were performed to verify the theoretical analysis presented. The experimental results show that the proposed PCA algorithm can converge to the Nash equilibrium very quickly, effectively reducing the peak value and increasing profit. [ABSTRACT FROM AUTHOR]

This paper introduces a tomography-like method for assessing grape maturation. It analyses inner tissue spectra through point-of-measurement (POM) sensing. A multi-block hierarchical principal component analysis (MHPCA) algorithm was used for the spectral reconstruction of total grapes (skin, pulp, and seed). Two grape cultivars, Loureiro (white; n = 216) and Vinhão (red; n = 205) were measured at 12 dates after veraison (DAV). The reconstructed spectra showed no significant differences (p < 0.001) from the originals for both grapes. Loureiro had better statistical metrics (Person's correlation coefficient (r) values for: total grape: 0.99, skin: 1; pulp: 1, seed: 0.94) than Vinhão (r values for: total grape: 0.92, skin: 0.92; pulp: 0.95, seed: 0.95). Using self-learning artificial intelligence (SL-AI), the following parameters were predicted for both grapes: soluble solids content (%; MAPE <13%), puncture force (N; MAPE <29%), chlorophyll content (a.u.; MAPE <29%), and anthocyanin content (a.u.; MAPE <17%, Vinhão only). When comparing observed values with predicted skin, pulp, and seed spectra, Vinhão showed no statistical differences for most parameters, except pulp chlorophyll on one DAV in the final maturation stage. The same was done with the Loureiro cultivar. Although Loureiro mostly showed no statistical differences in assessed parameters across tissues and dates, variations were found in pulp and skin chlorophyll content and puncture force. This tomography-like approach based on tissue maturation can help viticulturists to access instant data on grape maturation, supporting informed decision-making and promoting more sustainable agricultural practices. [Display omitted] • Bi-directional spectral reconstruction predicts the skin, pulp, seed, and total grape. • Spectral reconstruction works for a white and red grape cultivar. • Reconstructed spectra can estimate SSC, chlorophyll, anthocyanin and puncture force. • Tomography-like can support viticultural decisions based on grape physiology. [ABSTRACT FROM AUTHOR]

Although studies dealing with adsorption/desorption (and/or retention/release) of pollutants present in environmental compartments is a classical field of research, recent papers are focusing on some weak points of investigations and publications within the area. In addition, an increasing number of works are being published related to new possibilities and alternatives in this kind of research works, many of them in relation to the use of artificial intelligence (AI). Considering the existence of eventual controversies, eventual mistakes, and the convenience of suggesting alternatives to go ahead in the future, in this work, after taking into account some relevant publications in the previous literature, a simple workflow is proposed as a kind of protocol to revise successive steps that could guide the direction to follow when programing research dealing with the retention/release of pollutants in soils and sorbent materials. [Display omitted] • Recent papers show weak points of publications on pollutants adsorption/desorption. • Increasing works on the use of artificial intelligence in this kind of research. • A protocol is proposed to guide programing research on pollutants retention/release. [ABSTRACT FROM AUTHOR]

Our paper advances the embryonic interest of combining the theoretical frameworks of sustainable career and career ecosystem into a sustainable career ecosystem theory by introducing Artificial Intelligence (AI) as a new actor, spotlighting the need for liminality of the relationship between an individual and career practitioner, and presenting a new conceptual model. We begin by providing a brief overview of sustainable career and career ecosystem theories, culminating in a recently proposed definition of a sustainable career ecosystem. Second, using this as our point of departure, we consider the theoretical perspectives for understanding a sustainable career ecosystem through (a) introducing AI as a new actor with the potential to disrupt and transform the (future) labor market and (b) making a case for the liminality of the individual and career practitioner relationship. Third, we consider various dimensions for analyzing a sustainable career ecosystem to offer a new conceptual model. We conclude with a future research agenda. Conceptual Paper. • Integration of sustainable career theory and career ecosystem theory into sustainable career ecosystem theory • Development of a model portraying the conceptualization of a sustainable career ecosystem • Positioning artificial intelligence as a new, disruptive, and transformational actor within a sustainable career ecosystem • Consideration of the liminality dimension of the interaction between individuals and career practitioners • Offering a future research agenda to advance sustainable career ecosystems [ABSTRACT FROM AUTHOR]

Teenagers are the future and hope of the country. Sports play an essential role in promoting their physical and mental health. In the new historical era, the impact of artificial intelligence on sports is deepening. The application of network technology in youth sports can improve sports efficiency in the traditional environment, effectively promote the autonomy of young students in learning sports, and positively impact the realization of sports life. Further, this paper incorporates reinforcement theory, associative learning, and self-determination theory to explore the factors influencing youth sports participation more deeply. Through experimental analysis, this paper has understood the factors, characteristics, and differences that affect youth sports participation, analyzed the relationship between various factors and youth sports participation, constructed a structural equation model suitable for promoting youth sports participation, and finally took the Unified Theory of Acceptance and Use of Technology (UTAUT) model with more substantial explanatory power as the theoretical basis, and added three research variables: perceived fun, situation and trust. To establish the influencing factors model of teenagers' continuous participation in sports. The experimental results show that the optimized UTAUT model provides optimization ideas and technical support for youth sports training innovation, its error rate is reduced by 5%, and the average variance index is higher than 21%. [ABSTRACT FROM AUTHOR]

Federated Learning coordinates many mobile devices to train an artificial intelligence model while preserving data privacy collaboratively. Mobile devices are usually equipped with totally different hardware configurations, leading to various training capabilities. At the same time, the distribution of the local training data is highly heterogeneous across different clients. Randomly selecting the clients to participate in the training process results in poor model performance and low system efficiency. In this paper, we propose GraphCS, a graph-based client selection framework for heterogeneity in Federated Learning. GraphCS first measures the distribution coupling across the clients via the model gradients. After that, it divides the clients into different groups according to the diversity of the local datasets. At the same time, it well estimates the runtime training capability of each client by jointly considering the hardware configuration and resource contention caused by the concurrently running apps. With the distribution coupling information and runtime training capability, GraphCS selects the best clients in order to well balance the model accuracy and overall training progress. We evaluate the performance of GraphCS with mobile devices with different hardware configurations on various datasets. The experiment results show that our approach improves model accuracy up to 45.69%. Meanwhile, it reduces communication and computation overhead 87.35% and 89.48% at best, respectively. Furthermore, GraphCS accelerates the overall training process up to 35×. • We observe that the distribution of the training data can be reflected by the gradient of a machine learning model. • We observe that the training capability is jointly affected by the hardware configuration and resource contention. • we design a framework selects the best clients with diverse local datasets and highest runtime training capability. • We prototype GraphCS on a Federated system consisting of mobile devices with heterogeneous hardware configurations. [ABSTRACT FROM AUTHOR]

This study aims to comprehensively review the use of graph neural networks (GNNs) for clinical risk prediction based on electronic health records (EHRs). The primary goal is to provide an overview of the state-of-the-art of this subject, highlighting ongoing research efforts and identifying existing challenges in developing effective GNNs for improved prediction of clinical risks. A search was conducted in the Scopus, PubMed, ACM Digital Library, and Embase databases to identify relevant English-language papers that used GNNs for clinical risk prediction based on EHR data. The study includes original research papers published between January 2009 and May 2023. Following the initial screening process, 50 articles were included in the data collection. A significant increase in publications from 2020 was observed, with most selected papers focusing on diagnosis prediction (n = 36). The study revealed that the graph attention network (GAT) (n = 19) was the most prevalent architecture, and MIMIC-III (n = 23) was the most common data resource. GNNs are relevant tools for predicting clinical risk by accounting for the relational aspects among medical events and entities and managing large volumes of EHR data. Future studies in this area may address challenges such as EHR data heterogeneity, multimodality, and model interpretability, aiming to develop more holistic GNN models that can produce more accurate predictions, be effectively implemented in clinical settings, and ultimately improve patient care. [Display omitted] [ABSTRACT FROM AUTHOR]

In recent years, the increase in the use of services in cloud, fog, edge, and IoT ecosystems has been very notable. On the one hand, environmental sustainability is affected by this type of ecosystem since it can produce a large amount of energy consumption which translates into CO2 emissions into the atmosphere. On the other hand, due to the COVID-19 pandemic, the use of these ecosystems has increased considerably. Thus, it is necessary to apply policies and techniques to maximize sustainability within these ecosystems. Some of these policies and techniques are those based on artificial intelligence. However, the current processing of these policies and techniques can also consume a lot of resources. From this perspective, this article aims to clarify whether the sustainability of cloud/fog/edge/IoT ecosystems is improved by the application of artificial intelligence. To do this, a systematic literature review is developed in this paper. In addition, a set of classifications of the analyzed works is proposed based on the different aspects related to these ecosystems, their sustainability, and the applicability of artificial intelligence to improve them. • The use of cloud, fog, edge, and IoT has increased drastically in the last few years. • It is necessary to apply policies to maximize the sustainability of AI. • This work clarifies whether the sustainability of these environments is improved by AI. • A SLR is performed to achieve this aim. • The application of AI to these environments is a newborn topic. [ABSTRACT FROM AUTHOR]

The extensive use of Deep Neural Networks (DNNs) encourages people to design domain-specific hardware called Artificial Intelligence (AI) processors. The novel hardware makes optimizations challenging without a proper performance model that reveals working details and performance implications. This paper presents a performance model, Verrocchio, for Huawei DaVinci AI Core, which predicts the execution time of real-world DaVinci kernels. We propose specially-crafted micro-benchmarks to identify contention source, runtime behaviors, and bandwidth sharing, which significantly determine performance. Since DaVinci Core adopts a binary semaphore mechanism for synchronization, Verrocchio views each instruction as a discrete event and manages its execution time based on the programming logic. For evaluation, Verrocchio achieves average error rates of 2.62% and 2.30% in sample kernels for single-core and double-core execution. We demonstrate an optimizing process of matrix multiplications with Verrocchio, achieving speedups of 1.70× for operators and 1.53× for applications and error rates of 5.06% and 5.25%. • Detailed dissections of Huawei DaVinci AI Core, a novel AI processor. • Benchmarking the DaVinci Core bandwidth contention, the key performance factor. • Performance model for accurate execution time prediction of kernel program. • Demonstration of DaVinci kernel optimization and prediction accuracy evaluation. [ABSTRACT FROM AUTHOR]

Federated Learning is designed for multiple mobile devices to collaboratively train an artificial intelligence model while preserving data privacy. Instead of collecting the raw training data from mobile devices to the cloud, Federated Learning coordinates a group of devices to train a shared model in a distributed manner with the training data located on the devices. However, unbalanced data distribution and heterogeneous hardware configurations across different devices badly hurts the performance of collaborative model and severely impacts the overall training progress. Thus, a framework that can well balance the model accuracy and the training progress is urgently required. In this paper, we propose PAGroup, a privacy-aware grouping framework for high-performance Federated Learning. PAGroup intelligently divides the participating clients into different groups through carefully analyzing the privacy requirement of the training data and the solid social relationship of the participating clients. After that, PAGroup conducts data shaping and capability-ware training in order to improve the model performance while accelerating the overall training process. We evaluate the performance of PAGroup with both simulation and hardware testbed. The evaluation results show that PAGroup improves model accuracy up to 21%. Meanwhile, it decreases 81% communication overhead, 29% computation cost and 84% wall-clock time at best comparing with the baselines. • Specific private data that required by a learned model can be exchanged daily in certain social community. • PAGroup can share private date within task-specific groups organized according to strongly connected social community. • Training devices in a group can be viewed as a "super" client to compatibility to existing optimization algorithms. • The impact of local data scale, diversity, and label distribution to the model performance has been theoretically proved. [ABSTRACT FROM AUTHOR]

It is essential for implementing processing systems of edge computing, internet of things (IoT) and wireless multimedia sensor networks (WMSN) to use low-power parallel and distributed architectures with high speed and low power consumption. Most of the artificial intelligence and machine learning applications need to be executed on efficient distributed architectures with multicast support. In this paper, TLAM, a high-performance and low-cost NoC architecture is proposed. TLAM stands for Two-Level network-on-chip Architecture with Multicast support and includes a hybrid path/tree based multicast routing algorithm and a specific two-level mesh topology. The routing algorithm of TLAM is basically working according to the Hamiltonian paths. In the proposed architecture, the topology is partitioned and the two-level links provide an efficient infrastructure for low-cost and low-latency transmission of multicast and broadcast messages between partitions. In TLAM, in addition to multicasting routing algorithm, hardware components for handling multicasting packets are designed in order to achieve performance improvements. The goal is to improve the efficiency and performance, especially latency, while handling both unicast and multicast packets. Experimental evaluations including network-level and router-level analysis with different configurations under various traffic patterns were performed. The evaluations in terms of latency, throughput, area and power consumption, indicate that TLAM provides higher performance, especially for dense multicasting and broadcasting, in comparison with the existing architectures. • Exploration the potential of hybrid path/tree based multicasting by designing an efficient routing algorithm for high traffic NoCs. • Planning a partition-based two-level mesh topology, and connecting the partitions through second level links for multicast messages. • Designing the low-power and low-cost routers that support multicasting. • Conducting cycle-accurate network-level and router-level simulations of the propose architecture. • Analyzing the proposed design through extensive experiments considering different NoC configurations, various numbers of multicast packets for different numbers of destinations under both synthetic and real packet distributions. [ABSTRACT FROM AUTHOR]

[Display omitted] To evaluate the technical feasibility and potential value of a digital assistant that prompts intensive care unit (ICU) rounding teams to use evidence-based practices based on analysis of their real-time discussions. We evaluated a novel voice-based digital assistant which audio records and processes the ICU care team's rounding discussions to determine which evidence-based practices are applicable to the patient but have yet to be addressed by the team. The system would then prompt the team to consider indicated but not yet delivered practices, thereby reducing cognitive burden compared to traditional rigid rounding checklists. In a retrospective analysis, we applied automatic transcription, natural language processing, and a rule-based expert system to generate personalized prompts for each patient in 106 audio-recorded ICU rounding discussions. To assess technical feasibility, we compared the system's prompts to those created by experienced critical care nurses who directly observed rounds. To assess potential value, we also compared the system's prompts to a hypothetical paper checklist containing all evidence-based practices. The positive predictive value, negative predictive value, true positive rate, and true negative rate of the system's prompts were 0.45 ± 0.06, 0.83 ± 0.04, 0.68 ± 0.07, and 0.66 ± 0.04, respectively. If implemented in lieu of a paper checklist, the system would generate 56% fewer prompts per patient, with 50%±17% greater precision. A voice-based digital assistant can reduce prompts per patient compared to traditional approaches for improving evidence uptake on ICU rounds. Additional work is needed to evaluate field performance and team acceptance. [ABSTRACT FROM AUTHOR]

• Firstly, we present a new GNN-based model to predict NPI, where NPI's prediction problem is transformed into a link prediction problem. • Secondly, we propose a label trick mechanism in the ncRNA–protein bipartite graph to distinguish heterogeneous nodes ncRNA and protein, which could alleviate the problem of over-coupling. • Thirdly, we propose a self-attention operation that preserves the topology of the input graph during pooling. • Furthermore, We propose a community detection mechanism in the ncRNA–protein bipartite graph, which could represent the cluster (community). ownership relationship of the nodes and to further optimize the node embedding. Non-coding RNA (ncRNA) s play an considerable role in the current biological sciences, such as gene transcription, gene expression, etc. Exploring the ncRNA–protein interactions(NPI) is of great significance, while some experimental techniques are very expensive in terms of time consumption and labor cost. This has promoted the birth of some computational algorithms related to traditional statistics and artificial intelligence. However, these algorithms usually require the sequence or structural feature vector of the molecule. Although graph neural network (GNN) s has been widely used in recent academic and industrial researches, its potential remains unexplored in the field of detecting NPI. Hence, we present a novel GNN-based model to detect NPI in this paper, where the detecting problem of NPI is transformed into the graph link prediction problem. Specifically, the proposed method utilizes two groups of labels to distinguish two different types of nodes: ncRNA and protein, which alleviates the problem of over-coupling in graph network. Subsequently, ncRNA and protein embedding is initially optimized based on the cluster ownership relationship of nodes in the graph. Moreover, the model applies a self-attention mechanism to preserve the graph topology to reduce information loss during pooling. The experimental results indicate that the proposed model indeed has superior performance. [ABSTRACT FROM AUTHOR]

Over the last two decades, research in hyperspectral imaging has been increasing and its use in horticulture is expected to be spreading in the coming years. The emerging techniques are currently gaining interest of the research community. However, there are still challenges to the applicability. In this review we demonstrate that hyperspectral imaging can be used as an effective tool for fruit, vegetables and mushrooms in assessing quality parameters related to well defined variables that can be analysed in the laboratory, as well as complex properties such as maturity, ripeness, detection of biotic defects, physiological disorders, mechanical damages, and sensory quality. Therefore, this paper starts by giving an overview of the quality concept of produce, measuring principles, theory and analysis of hyperspectral imaging systems. Then, emerging techniques to monitor and assess quality parameters, both pre- and postharvest, are described, as well as applications of these are reviewed and discussed. Afterwards, this review proceeds by illustrating the current and potential use of artificial intelligence and its subdomains, machine learning and deep learning, for hyperspectral imaging analysis in horticulture. Lastly, some challenges and considerations for future research are highlighted, including improvement of data availability, possible solutions for an improved integration of artificial intelligence and the transfer of knowledge from research parameters to parameters relevant for industrial stakeholders. • Hyperspectral imaging is an effective tool for in assessing quality parameters. • The most abundantly used wavelengths are 601–850 nm, used in over 50% of studies. • PLSR and SVM are most commonly used models for machine learning. • Deep learning in combination with hyperspectral data is unexplored in horticulture. • Improving availability, AI knowledge transfer and standardisation is recommended. [ABSTRACT FROM AUTHOR]

Facial expression recognition technology has a brilliant prospect in applying the Internet of Things systems. Concerning the limited hardware computing capability and high real-time processing requirements, this paper proposes a lightweight emotion analysis system based on edge computing, which could be deployed on edge devices. To further improve the accuracy of dimensional emotion analysis, we propose a modified network structure of MobileNetV3 to measure the intensity of dimensional emotion. The optimization scheme includes introducing the improved efficient channel attention mechanism and the feature pyramid network, adjusting the structure of the model, and optimizing the loss function. Furthermore, the system uses Intel's OpenVINO toolkit to make the model more suitable for stable operation and provides an operable human–computer interaction interface. The experimental results show that the system has the advantages of few parameters, high recognition accuracy, and low latency. The optimized network size is reduced by 67% compared with the original, the root mean square error of potency and activation is 0.413 and 0.389, and the latency is up to 9ms in Myriad. This work meets the requirements of practical applications and has essential significance with the demand for continuous dimensional emotion analysis. The source code is available at https://github.com/SCNU-RISLAB/Lightweight-and-Continuous-Dimensional-Emotion-Analysis-System-of-Facial-Expression-Recognition/. [ABSTRACT FROM AUTHOR]

This paper presents a new approach for modelling and predicting the evolution of moisture content (MC) in couscous grains during drying using artificial neural networks. Couscous grains with an initial moisture content of 42 ± 1% were dried in an oven using the gravimetric method. Experimental data on moisture content versus drying time were obtained for nine groups of measurements at relative humidities of 7, 32 and 75% at drying temperatures of 30, 40 and 50°C. Different parameters of the ANN model, such as three learning algorithms and four activation functions, as well as the number of neurons in the hidden layer, were tested and evaluated using the correlation coefficient (R2) and the mean squared error (MSE) to find the optimal structure without overfitting. The best performing ANN structure consisted of the log sigmoid function in the hidden layer, with 24 neurons and the linear function in the output layer, trained by the Levenberg-Marquardt backpropagation algorithm. A significant agreement of predictions with experimental data was observed, with an R2 value of 0.99999 and an MSE value of 8.2173 ×10−4. • New ANN model for predicting the moisture content of couscous grains. • Various ANN structures and parameters assessed with MSE and R2. • This ANN excels in predicting couscous moisture with temperature and humidity. [ABSTRACT FROM AUTHOR]

The latest advances in artificial intelligence deep learning models are unprecedented. A wide spectrum of application areas is now thriving thanks to available massive training datasets and gigantic complex neural network models. Those two characteristics demand outstanding computing power that only advanced computing platforms can provide. Therefore, distributed deep learning has become a necessity in capitalizing on the potential of cutting-edge artificial intelligence. Two basic schemes have emerged in distributed learning. First, the data-parallel approach, which aims at dividing the training dataset into multiple computing nodes. Second, the model-parallel approach, which splits layers of a model into several computing nodes. Each scheme has its upsides and downsides, particularly when running on large machines that are susceptible to soft errors. Those errors occur as a consequence of several factors involved in the manufacturing process of current electronic components of supercomputers. On many occasions, those errors are expressed as bit flips that do not cause the whole system to crash, but generate wrong numerical results in computations. To study the effect of soft error on different approaches for distributed learning, we leverage checkpoint alteration, a technique that injects bit flips on checkpoint files. It allows researchers to understand the effect of soft errors on applications that produce checkpoint files in HDF5 format. This paper uses the popular deep learning PyTorch tool on two distributed-learning platforms: one for data-parallel training and one for model-parallel training. We use well-known deep learning models with popular training datasets to provide a picture of how soft errors challenge the training phase of a deep learning model. [ABSTRACT FROM AUTHOR]

Indole-3-acetic acid (IAA) derived from Actinobacteria fermentations on agro-wastes constitutes a safer and low-cost alternative to synthetic IAA. This study aims to select a high IAA-producing Streptomyces -like strain isolated from Lake Oubeira sediments (El Kala, Algeria) for further investigations (i.e., 16S rRNA gene barcoding and process optimization). Subsequently, artificial intelligence-based approaches were employed to maximize IAA bioproduction on spent coffee grounds as high-value-added feedstock. The specificity was the novel application of the Limited-Memory Broyden-Fletcher-Goldfarb-Shanno Box (L-BFGS-B) optimization algorithm. The new strain AW08 was a significant producer of IAA (26.116 ± 0.61 μg/mL) and was identified as Streptomyces rutgersensis by 16S rRNA gene barcoding and phylogenetic inquiry. The empirical data involved the inoculation of AW08 in various cultural conditions according to a four-factor Box Behnken Design matrix (BBD) of Response surface methodology (RSM). The input parameters and regression equation extracted from the RSM-BBD were the basis for implementing and training the L-BFGS-B algorithm. Upon training the model, the optimal conditions suggested by the BBD and L-BFGS-B algorithm were, respectively, L-Trp (X1) = 0.58 %; 0.57 %; T° (X2) = 26.37 °C; 28.19 °C; pH (X3) = 7.75; 8.59; and carbon source (X4) = 30 %; 33.29 %, with the predicted response IAA (Y) = 152.8; 169.18 μg/mL). Our findings emphasize the potential of the multifunctional S. rutgersensis AW08, isolated and reported for the first time in Algeria, as a robust producer of IAA. Validation investigations using the bioprocess parameters provided by the L-BFGS-B and the BBD-RSM models demonstrate the effectiveness of AI-driven optimization in maximizing IAA output by 5.43-fold and 4.2-fold, respectively. This study constitutes the first paper reporting a novel interdisciplinary approach and providing insights into biotechnological advancements. These results support for the first time a reasonable approach for valorizing spent coffee grounds as feedstock for sustainable and economic IAA production from S. rutgersensis AW08. [Display omitted] • Multifunctional Streptomyces rutgersensis AW 08 represents a novel viable source for indole-3-acetic acid. • Spent coffee grounds can substitute laboratory-grade substrates for Streptomyces fermentation. • L-BFGS-B algorithm implementation with RSM-BBD achieves fast and effective bioprocess optimization. • Artificial intelligence-driven approaches increase indole-3-acetic acid yields up to fivefold. [ABSTRACT FROM AUTHOR]

Flood modelling and forecasting can enhance our understanding of flood mechanisms and facilitate effective management of flood risk. Conventional flood hazard and risk assessments usually consider one driver at a time, whether it is ocean, fluvial or pluvial, without considering the compound nature of flood events. In this paper, we developed a novel approach for modelling and forecasting compound coastal-fluvial floods using a two-step framework. In step one, a hydrodynamic model is used to simulate floodwater propagation; while in step two, machine learning (ML) models are used to generate flood forecasts. The architecture of hydrodynamic-ML forecasting system incorporates a hydrodynamic model covering a specific domain, with individual ML models trained for each pixel. In total 7 ML models including: Support Vector Regression (SVR), Support Vector Machine (SVM), Radial Basis Function (RBF), Linear Regression (LR), Gaussian Process Regression (GPR), Decision Tree (DT), and Artificial Neural Network (ANN) were applied in this study. Forecasting compound floods is achieved using two sets of inputs: timeseries of river discharges in the upstream fluvial section and downstream ocean water levels in the coastal areas. The accuracy of the flood forecasting system is demonstrated for Cork City, Ireland; and modelling performance was evaluated using several statistical tools. Results show that the proposed models can provide reliable estimates of flood inundation and associated water depths. Overall, the RBF model exhibits the best performance. Despite the complexity of compound multi-driver floods, this study shows that the coupled hydrodynamic-ML approach can forecast coastal-fluvial flood with limited hydraulic and hydrological input data. This system overcomes the limitations of traditional hydrodynamic model-based systems where trade-offs between the always competing numerical model accuracy and computational time prohibit the model to be used for short-term flood forecasting. Once trained, the ML component of the coupled system can perform flood forecasting in near real-time, potentially integrating into a flood early warning system. Accurate flood forecasting has a wide range of positive societal impacts, including improved flood preparedness, increased confidence, better resource allocation, reduced flood damage, and potentially even flood prevention. • Novel technique to improve compound coastal-fluvial flood modelling. • Flood modelling technique based on seven Machine Learning methods. • Approach is successfully implemented, and results are evaluated over Cork, Ireland. • The results demonstrate minimal quantitative errors for the proposed models. [ABSTRACT FROM AUTHOR]

• Integration of Explainable Artificial Intelligence (XAI) models on fiber optic SPR sensors. • Insights into the intricate patterns and features learned by the model are attained. • A comparative analysis of XAI models, namely SHAP and LIME is carried out. • LIME and SHAP methods jointly find that metal layer thickness has considerably greater influence than wavelength on sensing performance. • This finding is consistent with the concept of radiation damping related to fiber optic SPR sensor. In recent years, there has been a significant surge in interest surrounding the utilization of machine learning techniques to optimize the performance of fiber optic sensors. This paper delves into the integration of machine learning methodologies, with a specific focus on Explainable Artificial Intelligence (XAI) models, within the domain of fiber optic surface plasmon resonance (SPR) sensors. We implement XAI techniques into a previously-trained Gaussian Process Regression (GPR) model to gain insights into the intricate patterns and features learned by the model, thereby providing valuable insights into its decision-making process. Furthermore, we present a comparative analysis of XAI models, namely SHAP and LIME , to elucidate the individual impact of light wavelength (λ) and metal layer thickness (d m) on the enhancement of sensor's figure of merit (FOM). The results obtained from both LIME and SHAP methods lead to a common finding that compared to λ, it is the value of d m that possesses a significantly greater influence on achieving maximum FOM. Furthermore, this finding is in complete agreement with the concept of radiation damping related to fiber optic SPR sensor structures. This study has the potential to contribute to the refinement of design and fabrication processes related to fiber optic SPR sensor for a wide variety of applications. [ABSTRACT FROM AUTHOR]

In recent years there has been increased use of machine learning (ML) techniques within mathematics, including symbolic computation where it may be applied safely to optimise or select algorithms. This paper explores whether using explainable AI (XAI) techniques on such ML models can offer new insight for symbolic computation, inspiring new implementations within computer algebra systems that do not directly call upon AI tools. We present a case study on the use of ML to select the variable ordering for cylindrical algebraic decomposition. It has already been demonstrated that ML can make the choice well, but here we show how the SHAP tool for explainability can be used to inform new heuristics of a size and complexity similar to those human-designed heuristics currently commonly used in symbolic computation. [ABSTRACT FROM AUTHOR]

This paper focuses on predicting predawn leaf water potential through a self-learning artificial intelligence (SL-AI) algorithm, a novel spectral processing algorithm that is based on the search for covariance modes, providing a direct relationship between spectral information and plant constituents. The SL-AI algorithm was applied in a dataset containing 847 observations obtained with a handheld hyperspectral spectroradiometer (400–1010 nm), structured as: three grapevine cultivars (Touriga Nacional, Touriga Franca and Tinta Barroca), collected in three years (2014, 2015 and 2017), in two test sites in the renowned Douro Wine Region, northeast of Portugal. The Ψ pd SL-AI quantification was tested both in regressive (R2 = 0.97, MAPE = 18.30%) and classification (three classes; overall accuracy = 86.27%) approaches, where the radiation absorption spectrum zones of the chlorophylls, xanthophyll and water were identified along the vegetative growth cycle. The dataset was also tested with Artificial Neural Networks with Principal Component Analysis (ANN-PCA) and Partial Least Square (PLS), which presented worse performance when compared to SL-AI in the regressive (ANN-PCA - R2 = 0.85, MAPE = 43.64%; PLS - R2 = 0.94, MAPE = 28.76%) and classification (ANN-PCA - overall accuracy: 72.37%; PLS - overall accuracy: 73.79%) approaches. The Ψ pd modelled with SL-AI demonstrated, through hyperspectral reflectance, a cause-effect of the grapevine's hydric status with the absorbance of bands related to chlorophyll, xanthophylls and water. This cause-effect interaction could be explored to identify cultivars and cultural practices, hydric, heating and lighting stresses. [Display omitted] • VIS-NIR spectroscopy was used to predict predawn leaf water potential (Ψ pd). • Benchmark of SL-AI against state-of-the-art chemometrics methods in modelling Ψ pd. • Chlorophyll, xanthophyll, and water absorbance bands related to the Ψ pd. • Cause-effect of plant response to abiotic stresses explained through VIS-NIR bands. [ABSTRACT FROM AUTHOR]

Based on MPP database, we have conducted research and investigation on big data processing and analysis, and provided a good solution for big data. However, due to the large data storage requirements in the local file system may not be fully met. Therefore, the storage and management of big data, reliability, availability and scalability are more and more adopted by distributed file system. Therefore, MPP database based on distributed file system has become a research hotspot. In this paper, in the network control, through the control algorithm based on two kinds of artificial intelligence system, through the fuzzy PID controller to reduce the impact of system control performance. The dynamic BP network error correction and predictive PID controller are introduced into the system to reduce the influence. College Students' mental health education is an important part of the current training. Rapid and accurate online evaluation and analysis of College Students' mental health can provide reference for college students and effective help for the follow-up activities and consultation of the school, At the same time, it provides some basis for the balance of psychological scale and the revision of item setting. In this paper, MPP database and artificial intelligence system are introduced into the application of mental health assessment, which can effectively assess the mental health of college students, and thus promote the continuous development of assessment application. • MPP database based on distributed file system has become a research hotspot. • The dynamic BP network error correction and predictive PID controller are introduced into the system. • MPP database and artificial intelligence system are introduced into the application of mental health. [ABSTRACT FROM AUTHOR]

[Display omitted] • This paper reviews clinical decision system architectures utilizing a combination of rule-based reasoning and machine learning to identify recurrent architecture archetypes and motivations of their authors. • Five distinct architecture archetypes were identified: Rules are Embedded in ML architecture (REML), ML pre-processes input data for Rule-Based inference (MLRB), Rule-Based method pre-processes input data for ML prediction (RBML), Rules influence ML training (RMLT), Parallel Ensemble of Rules and ML (PERML). • Parallel ensembles that leverage the benefits of machine learning and rule-based reasoning are rare in the present literature and offer a promising avenue for further research and exploration. As the application of Artificial Intelligence (AI) technologies increases in the healthcare sector, the industry faces a need to combine medical knowledge, often expressed as clinical rules, with advances in machine learning (ML), which offer high prediction accuracy at the expense of transparency of decision making. This paper seeks to review the present literature, identify hybrid architecture patterns that incorporate rules and machine learning, and evaluate the rationale behind their selection to inform future development and research on the design of transparent and precise clinical decision systems. PubMed, IEEE Explore, and Google Scholar were queried in search for papers from 1992 to 2022, with the keywords: "clinical decision system", "hybrid clinical architecture", "machine learning and clinical rules". Excluded articles did not use both ML and rules or did not provide any explanation of employed architecture. A proposed taxonomy was used to organize the results, analyze them, and depict them in graphical and tabular form. Two researchers, one with expertise in rule-based systems and another in ML, reviewed identified papers and discussed the work to minimize bias, and the third one re-reviewed the work to ensure consistency of reporting. The authors screened 957 papers and reviewed 71 that met their criteria. Five distinct architecture archetypes were determined: Rules are Embedded in ML architecture (REML) (most used), ML pre-processes input data for Rule-Based inference (MLRB), Rule-Based method pre-processes input data for ML prediction (RBML), Rules influence ML training (RMLT), Parallel Ensemble of Rules and ML (PERML), which was rarely observed in clinical contexts. Most architectures in the reviewed literature prioritize prediction accuracy over explainability and trustworthiness, which has led to more complex embedded approaches. Alternatively, parallel (PERML) architectures may be employed, allowing for a more transparent system that is easier to explain to patients and clinicians. The potential of this approach warrants further research. A limitation of the study may be that it reviews scientific literature, while algorithms implemented in clinical practice may present different distributions of motivations and implementations of hybrid architectures. [ABSTRACT FROM AUTHOR]

Recent advances in Artificial Intelligence (AI) and Machine Learning (ML) brought light on the need for explainability in multiple domains (e.g., healthcare, finance, justice, and recruiting). Explainability or Explainable AI (XAI) can be defined as everything that makes AI more understandable to human beings. However, XAI features may vary according to the AI algorithm used. Beyond XAI features, different AI algorithms vary in terms of speed, performance, and costs associated with training/running models. Knowing when to choose the right algorithm for the task at hand, therefore, is fundamental in multiple AI systems, for instance, AutoML and AutoAI. In this paper, we propose a method to analyze patterns of high-granularity user interface (UI) events (i.e., mouse, keyboard, and additional custom events triggered on the millisecond scale) to predict when users will interact with UI elements that provide explainability for the AI in place. In this context, this paper presents: (1) a user study involving 37 participants (7 in the pilot phase and 30 in the main experiment phase) in which people performed a task of reporting a bug using a text form associated with an AI data quality meter and its XAI UI element and (2) an approach to model micro behavior using n o d e 2 v e c to predict when the interaction with XAI UI element will occur. The proposed approach uses a rich dataset (approximately 129k events) and combines n o d e 2 v e c and a Logistic Regression classifier. Results obtained show we have obtained an event-by-event prediction of the interaction with XAI with an average F-score of 0.90 (σ = 0. 06). From the presented results, one expects to support researchers in the realm of UI personalization to consider high-granularity interaction data when predicting the need for XAI while users are interacting with AI model outputs. • A study detailing different behaviors while users interact with explainability. • A user model approach to predict interaction with explainability. • The proposed approach obtained an average F-score of 0.90. • Groups do not differ in terms of task time nor event stream length. • An extension for XAI taxonomy encompassing interaction with XAI. [ABSTRACT FROM AUTHOR]

• Propose an energy efficient EEC-SDTP technique for Intelligent IoT Edge systems. • The EEC-SDTP technique aims to select CHs and secure routes for data transmission. • Designs a novel OCGOC technique for proper CH selection and cluster construction. • Derive a SRP-QSPO technique to achieve secure data transmission in IoT edge systems. • Validate the performance of the EEC-SDTP technique interms of different measures. In the past decade, the Internet of Things (IoT) becomes essential in consumer and industrial applications. The accessibility of high bandwidth Internet connection particularly with the arrival of robust 5G networks rises to innovative IoT solutions such as smart city, automobiles, industry 4.0, etc. IoT analytics represents edge computing as a term commonly employed for defining intelligent computational resources placed closer to the source of data generation. Despite the benefits of the IoT edge systems, security and energy efficiency remains major challenging issues. With this motivation, this paper presents an energy-efficient clustering based secure data transmission protocol (EEC-SDTP) for Intelligent IoT Edge systems. The goal of the EEC-SDTP technique is to select an appropriate set of cluster heads (CHs) and optimally secure routes for data transmission in the network. The proposed model involves oppositional chaos game optimization-based clustering (OCGOC) technique for proper CH selection and cluster construction. Besides, a trust-based model is designed to determine the trustworthiness of the node in the IoT edge systems. The proposed OCGOC technique derives a fitness function utilizing 3 input parameters like trust level, distance to neighbors, and energy. Finally, a trust-based secure routing protocol using the quantum sand piper optimization (SRP-QSPO) technique is employed to derive routes for secure data transmission. For examining the better efficiency of the proposed EEC-SDTP algorithm, an extensive group of experimentations were performed and the outcomes are investigated under several performance measures. The experimental outcomes highlighted the improved efficiency of the proposed method over the other related techniques. [ABSTRACT FROM AUTHOR]

The edge computing (EC) systems for Internet of Things (IoT) can bring out low latency, high reliability, distributed intelligence, and network bandwidth savings to industrial real-time applications. However, limited computing and processing capabilities of edge devices remains to be difficult to meet complex data processing and artificial intelligence (AI) analysis requirements for diverse services. Besides, the scarcity of wireless spectrum resources in harsh industrial environment makes the interference between devices more serious. To address these challenges, this paper proposes an adaptive distributed collaborative anti-interference optimization scheme for IoT-edge system. Firstly, the EC system model is established, and the model of link failure probability is derived theoretically. Then, an Occupy-Interference Mitigation (O-IM) algorithm based on full-frequency multiplexing is proposed. The algorithm combines adaptive full-frequency multiplexing and interference mitigation to reduce the dependence of the reliable collaboration on bandwidth and signal-to-noise ratio (SNR). In addition, an anti-interference algorithm based on Interference Mitigation and inter-cellfrequency multiplexing Interference Avoidance (IM-IA) is proposed to balance bandwidth and SNR. This algorithm adopts interference cancellation scheme in the broadcasting phase, and adopts orthogonal frequency division in the collaboration phase. Extensive simulation results on Mininet platform verify that the proposals can obtain lower failure probability, and are less than traditional solutions in transmitting power, bandwidth, and SNR requirements. Moreover, the O-IM is suitable for low power scenarios, while the IM-IA is more suitable for high power case. • The edge computing system model for IoT is constructed. • An Occupy-Interference Mitigation (O-IM) algorithm is proposed. • An anti-interference algorithm based on IM-IA is proposed. • The proposals obtain lower failure probability and channel quality requirements. [ABSTRACT FROM AUTHOR]

Cloud Computing is defined as a set of software and hardware that are used together for delivering different kinds of cloud services based on user's demand. Cloud computing secures a major role in the Information Technology (IT) industry for accessing the services at any place around the world. On the other hand, there are increasing vulnerabilities and threats in the cloud environment due to the rise in popularity and demands in cloud computing services. Data privacy and integrity are the major issues in cloud computing while storing data in various geographical locations. So, it is necessary to consider the data privacy and integrity factors in a cloud computing environment. It is difficult to construct a common platform to interact in the cloud environment. So, it is essential for implementing the security solutions, which need to provide confidentiality while exchanging the data. The main intention of this paper is to design and develop a novel artificial intelligence approach for handling the privacy preservation problem in the cloud sector. Through the process of data sanitization, the sensitive data is hidden, so that it cannot be accessed by unauthorized users. To perform the sanitization process, the heuristic-based key generations play a vital role, and here, it is solved by considering a multi-objective function with constraints like the "degree of modification, hiding ratio, and information preservation ratio". This multi-objective problem is solved by the adoption of novel Probability Switch searched Butterfly-Moth Flame Optimization (PS-BMFO). Finally, the restoration is also performed by the same PS-BMFO-based key generation. The analysis results confirm that the suggested model preserves privacy and ensures the integrity of the user's data against unauthorized parties. From the experimental analysis, the proposed PS-BMFO gives 12.5%, 10%, 30.7%, and 12.5% enriched than GWO, JA, MFO, and BOA, respectively. Therefore, the statistical analysis shows that the developed data privacy preservation model with suggested PS-BMFO performs better than the other conventional algorithms. • Develop a hybrid PS-BMFO algorithm for securing the information sharing along with ensuring the privacy access of cloud data. • Introduce the hybrid PS-BMFO for securing the information sharing and ensuring the privacy access of cloud data. • Implement the privacy preservation techniques with data sanitization and data restoration processes using PS-BMFO. • Use hybrid PS-BMFO algorithm for minimizing the hiding ratio, information preservation ratio and degree of modification. • Analyze the efficiency of proposed model with developed PS-BMFO by comparing with existing meta-heuristic algorithms. [ABSTRACT FROM AUTHOR]

Wireless-optical broadband access network (WOBAN) or fiber-wireless (FiWi) is a robust and cost-effective solution to provide high bandwidth network connectivity to mobile subscribers. Optical network unit (ONU) placement is a topological optimization problem that plays a vital role in effective deployment of FiWi network. Recently, some artificial intelligence-based approaches for positioning ONUs were proposed and generated promising results. However, the reported performance in the literature is far from the ideal state. This issue emphasizes the necessity of introducing new placement methods or improving the performance of existing methods. This paper presents a hybrid clustering algorithm for finding the optimal position of multiple ONUs. The proposed backtracking K -harmonic means (BKHM) is an integration of K -harmonic means (KHM) and a backtracking search algorithm (BSA). BKHM not only efficiently specifies the optimal positions of ONUs but also significantly decreases the implementation cost. To demonstrate the performance of the BKHM, we evaluate it on well-known placement benchmarks compared with BSA, KHM, whale optimization algorithm (WOA), enhanced backtracking search (EBS), harris hawks optimization (HHO), and chaotic local search-based levy flight distribution (CLF) algorithms. The experimental results and statistical analysis proved that the proposed BKHM algorithm outperformed its counterparts on most benchmarks in terms of cost optimization, convergence accuracy, and solution quality. • Proposing a hybrid clustering algorithm: The proposed not only solves the slow convergence of the BSA algorithm but also helps the KHM to escape from local optimal. • Using the BKHM for solving ONU placement problem: The BSA algorithm is suitable for solving permutation problems, therefore combining it with KHM can work better than other algorithms for the problem of ONU placement. • Improving the performance of the existing ONU placement methods. Simulation results confirm the superiority of the BKHM compared with other methods. [ABSTRACT FROM AUTHOR]

Recent advances in artificial intelligence (AI) have inspired researchers to explore machine learning (ML)-based optimization and reverse design techniques for photonic crystal fibers (PCFs). These studies often seek to improve model generalization, particularly for data that the model has not previously encountered. Traditional centralized training methods are challenging for devices with limited resources, as they rely on aggregating expansive datasets, which is hindered by constraints in storage capacity and communication efficiency. This paper introduces an innovative distributed framework for optimizing PCF parameters, utilizing decentralized training to amalgamate knowledge across various institutions while maintaining data privacy. Each institution develops a lightweight neural network using a small subset of local data, contributing to the construction of a collective and robust global model. This approach is advantageous for both internal and external applications in PCF engineering. Rigorous empirical experiments conducted with real-world PCF optimization data substantiate the efficacy and benefits of the proposed framework. This framework shows promise in achieving an equilibrium between data protection and resource efficiency, offering a novel platform for the reverse design of microstructured optical fibers. [ABSTRACT FROM AUTHOR]

• In-depth review of deep learning methods for seizure video analysis (last 7 years). • Introducing a multi-stream framework to promote AI adoption in Epilepsy. • Unveiling solutions in AI for seizure analysis challenges overlooked in prior reviews. Seizure events can manifest as transient disruptions in the control of movements which may be organized in distinct behavioral sequences, accompanied or not by other observable features such as altered facial expressions. The analysis of these clinical signs, referred to as semiology, is subject to observer variations when specialists evaluate video-recorded events in the clinical setting. To enhance the accuracy and consistency of evaluations, computer-aided video analysis of seizures has emerged as a natural avenue. In the field of medical applications, deep learning and computer vision approaches have driven substantial advancements. Historically, these approaches have been used for disease detection, classification, and prediction using diagnostic data; however, there has been limited exploration of their application in evaluating video-based motion detection in the clinical epileptology setting. While vision-based technologies do not aim to replace clinical expertise, they can significantly contribute to medical decision-making and patient care by providing quantitative evidence and decision support. Behavior monitoring tools offer several advantages such as providing objective information, detecting challenging-to-observe events, reducing documentation efforts, and extending assessment capabilities to areas with limited expertise. The main applications of these could be (1) improved seizure detection methods; (2) refined semiology analysis for predicting seizure type and cerebral localization. In this paper, we detail the foundation technologies used in vision-based systems in the analysis of seizure videos, highlighting their success in semiology detection and analysis, focusing on work published in the last 7 years. We systematically present these methods and indicate how the adoption of deep learning for the analysis of video recordings of seizures could be approached. Additionally, we illustrate how existing technologies can be interconnected through an integrated system for video-based semiology analysis. Each module can be customized and improved by adapting more accurate and robust deep learning approaches as these evolve. Finally, we discuss challenges and research directions for future studies. [ABSTRACT FROM AUTHOR]

Encapsulated cell technology (ECT) is a targeted delivery method that uses the genetically engineered cells in semipermeable polymer capsules to deliver cytokines. Thus far, ECT has been extensively utilized in pharmacologic research, and shows enormous potentials in the treatment of posterior segment diseases. Due to the biological barriers within the eyeball, it is difficult to attain effective therapeutic concentration in the posterior segment through topical administration of drug molecules. Encouragingly, therapeutic cytokines provided by ECT can cross these biological barriers and achieve sustained release at the desired location. The encapsulation system uses permeable materials that allow growth factors and cytokines to diffuse efficiently into retinal tissue. Moreover, the ECT based treatment can be terminated timely when we need to retrieve the implant, which makes the therapy reversible and provides a safer alternative for intraocular gene therapy. Meanwhile, we also place special emphasis on optimizing encapsulation materials and enhancing preservation techniques to achieve the stable release of growth factors and cytokines in the eyeball. This technology holds great promise for the treatment of patients with dry AMD, RP, glaucoma and MacTel. These findings would enrich our understandings of ECT and promote its future applications in treatment of degenerative retinopathy. This review comprises articles evaluating the exactness of artificial intelligence-based formulas published from 2000 to March 2024. The papers were identified by a literature search of various databases (PubMed/MEDLINE, Google Scholar, Cochrane Library and Web of Science). [Display omitted] [ABSTRACT FROM AUTHOR]

• A new adaptive blind image watermarking technology, utilizing artificial intelligence (AI) and named AwDD, has been proposed for detecting color image deepfakes. • The AI technology used includes face detection, denoising autoencoder (DAE), and the gray wolf optimizer (GWO). • The proposed method employs mixed modulation (MM) with partly sign-altered mean value (PSAMV) to embed a set of coefficients, thereby increasing robustness against attacks while maintaining high image quality. • Blind adaptive deepfake detection (BADD) with the tamper detection mean value (TDMV) is applied to adaptively detect changes in the relative position when a face image is modified or deepfaked. • The GWO optimizes parameters to enhance imperceptibility and robustness, while the DAE improves the visual recognizability of the extracted watermark. This paper proposes a new adaptive blind watermarking technology for deepfake detection, which can embed deepfake detection information into the image and verify the image's authenticity without requiring additional information. The proposed scheme utilizes mixed modulation combined with partly sign-altered mean value to embed a set of coefficients that enhance robustness against attacks while maintaining high image quality. Additionally, blind adaptive deepfake detection technology with the tamper detection mean value is employed to detect relative positions adaptively, even when face images are slightly modified or deepfaked. To further improve the performance of the proposed scheme, a gray wolf optimizer is introduced to optimize parameters, and a denoising autoencoder is employed to facilitate the identification of extracted watermarks. This technology will adaptively embed watermark information while preserving the original face image, thereby maintaining the authenticity of the face in the image and verifying the owner of the image. The code is available at https://github.com/lyhsu01/AwDD. [ABSTRACT FROM AUTHOR]

Partially fake audio, a variant of deep fake that involves manipulating audio utterances through the incorporation of fake or externally-sourced bona fide audio clips, constitutes a growing threat as an audio forgery attack impacting both human and artificial intelligence applications. Researchers have recently developed valuable databases to aid in the development of effective countermeasures against such attacks. While existing countermeasures mainly focus on identifying partially fake audio at the level of entire utterances or segments, this paper introduces a paradigm shift by proposing frame-level systems. These systems are designed to detect manipulated utterances and pinpoint the specific regions within partially fake audio where the manipulation occurs. Our approach leverages acoustic features extracted from large-scale self-supervised pre-training models, delivering promising results evaluated on diverse, publicly accessible databases. Additionally, we study the integration of boundary and deepfake detection systems, exploring their potential synergies and shortcomings. Importantly, our techniques have yielded impressive results. We have achieved state-of-the-art performance on the test dataset of the Track 2 of ADD 2022 challenge with an equal error rate of 4.4%. Furthermore, our methods exhibit remarkable performance in locating manipulated regions in Track 2 of the ADD 2023 challenge, resulting in a final ADD score of 0.6713 and securing the top position. • Simultaneously detecting partially spoofed audio and locating the manipulated regions. • Exploring model integration techniques in identifying fake regions. • Investigating self-supervised pre-training models in partially spoofing detection. [ABSTRACT FROM AUTHOR]

Mobile devices (MDs) and applications are receiving extensive popularity and attracting significant attention. Mobile applications, especially for artificial intelligence (AI) applications, require powerful computation-intensive resources. Hence, running all the AI applications on a single MD introduces high energy consumption and application delay, as it has limited battery capacity and computation resources. Fortunately, the emerging edge-cloud computing (ECC) architecture pushes the computation resource to both the network edge and remote cloud to cope with challenging AI applications. Although the advantage of ECC greatly benefits various mobile applications, data security remains an important open issue in this scenario, which has not been well studied. This paper focuses on the profit maximization (PM) problem for security-aware task offloading in an ECC environment, i.e., considering the tasks from MDs with different service demands, edge nodes should decide them to be processed on the edge node or the remote cloud with a security guarantee. Specifically, we first construct the security model to measure the time overhead for each task under various scenarios. We then formulate the PM problem by jointly considering the security demand and deadline constraints of tasks. Finally, we propose a genetic algorithm-based PM (GA-PM) algorithm, the coding strategy of which considers the task execution location and execution order. Moreover, the crossover and mutation operations are implemented based on the coding strategy. Extensive simulation experiments with various parameters varying demonstrate that our GA-PM can achieve better performance than all the comparison algorithms. • The security model is built to measure the execution time of tasks under different parameters. • A genetic algorithm (GA)-based PM algorithm is proposed to implement task offloading and obtain optimal profit. • A coding strategy is devised by considering the tasks execution location and execution order. • Extensive simulation results demonstrate that GA-PM algorithm achieves the optimal profit. [ABSTRACT FROM AUTHOR]

The topic of algorithmic fairness is of increasing importance to the Human–Computer Interaction research community following accumulating concerns regarding the use and deployment of Artificial Intelligence-based systems. How we conduct research on algorithmic fairness directly influences our inferences and conclusions regarding algorithmic fairness. To better understand the methodological decisions of studies focused on people's perceptions of algorithmic fairness, we systematic analysed relevant papers from the CHI and FAccT conferences. We identified 200 relevant papers published between 1993 and 2022 and assessed their study design, participant sample, and geographical location of participants and authors. Our results highlight that studies are predominantly cross-sectional, cover a wide range of participant roles, and that both authors and participants are primarily from the United States. Based on these findings, we reflect on the potential pitfalls and shortcomings in how the community studies algorithmic fairness. • Mapping of CHI and FAccT papers on algorithmic fairness published 1993–2022. • Over one-third of papers miss details on participant locality and compensation. • Algorithmic fairness research largely restricted to US and other Western countries. • Limited cross-country collaboration, most study samples from a single country. • Remote studies are most common, longitudinal studies are relatively rare. [ABSTRACT FROM AUTHOR]

Existing knowledge distillation methods use large amount of bilingual data and focus on mining the corresponding knowledge distribution between the source language and the target language. However, for some languages, bilingual data is not abundant. In this paper, to make better use of both monolingual and limited bilingual data, we propose a new knowledge distillation method called Dual Knowledge Distillation (DKD). For monolingual data, we use a self-distillation strategy which combines self-training and knowledge distillation for the encoder to extract more consistent monolingual representation. For bilingual data, on top of the k Nearest Neighbor Knowledge Distillation (kNN-KD) method, a similar self-distillation strategy is adopted as a consistency regularization method to force the decoder to produce consistent output. Experiments on standard datasets, multi-domain translation datasets, and low-resource datasets show that DKD achieves consistent improvements over state-of-the-art baselines including kNN-KD. • Monolingual data can better help low-resource machine translation. • Dual Knowledge Distillation can better use bilingual and monolingual data. • Self-distillation can extract more consistent monolingual representation. • Consistency regularization can force the decoder to generate consistent output. [ABSTRACT FROM AUTHOR]

Document-level Relation Extraction(DocRE) aims to extract relations between entities from documents. In contrast to sentence-level relation extraction, it requires extracting semantic relations from multiple sentences. It is necessary to further improve the performance of the above algorithm in order to extract document-level relation. Therefore, the DocRE algorithms have to deal with more complex entity structure relationships and the need to unite semantic relationships between different sentences when reasoning about relationships between entities. The proposed algorithms fail to infer relationships between entities when dealing with complex entity structure relationships. In this paper, we propose an entity mentions deep attention framework that efficiently infers entity relationships through entity structure and contextual information. Firstly, a structural dependency module of entities is designed to achieve interaction between different mentions of the entity. Secondly, a deep contextual attention component proposed to enrich the semantic information between entities by entity-related contexts. Finally, we use a distance mapping component to solve the problem of entity pairs that are far away from each other. According to our implementation results, our model outperforms the state-ofthe-art models on three public datasets DocRED, DGA, and CDR. • We created dependencies for entities and various mentions to enable structural reasoning. • The deep contextual attention module is designed to enhance context semantic information and improve model reasoning. • A location mapping module enhances entity representations for accurate capture of remote entity relationships. • The experimental results demonstrate the superiority of our EMDA model. [ABSTRACT FROM AUTHOR]

The last decade has witnessed enormous improvements in science and technology, stimulating the growing demand for economic and cultural exchanges in various countries. Building a neural machine translation (NMT) system has become an urgent trend, especially in the low-resource setting. However, recent work tends to study NMT systems for low-resource languages centered on English, while a few studies focus on low-resource NMT systems centered on other languages such as Chinese. To achieve this, the low-resource multilingual translation challenge of the 2021 iFLYTEK AI Developer Competition provides the Chinese-centric multilingual low-resource NMT tasks, where participants are required to build NMT systems based on the provided low-resource samples. In this paper, we present the competition-winning system that leverages monolingual word embeddings data enhancement, bilingual curriculum learning, and contrastive re-ranking. In addition, a new Incomplete-Trust (In-trust) loss function is proposed to replace the traditional cross-entropy loss when training. The extensive experimental results demonstrate that the implementation of these ideas leads to better performance than other state-of-the-art methods. All the experimental codes are released at: https://github.com/WENGSYX/Low-resource-text-translation. • Focusing on Chinese low-resource neural translation for further development. • Presenting new ideas for translation by enhancing pre-trained language models. • Self-developing optimization loss to further improve translation quality. • The proposed system outperforms the existing translation engine. [ABSTRACT FROM AUTHOR]

• An automated brain parcellation tool, 'Auto-Parc', has been developed to identify cortical regions and subcortical structures, as well as the locations of their supplying vascular structures. • The tool identifies 58 regions using a deep, 3D, sliding-window, U-Net architecture. • The use of a single network for the entire image allows for rapid re-training of the network as needed. • The tool is used as the first step of a fully-automated fMRI analysis protocol for assessment of mTBIs. Functional MRI has emerged as a powerful tool to assess the severity of Post-concussion syndrome (PCS) and to provide guidance for neuro-cognitive therapists during treatment. The next-generation functional neuro-cognitive imaging protocol (fNCI2) has been developed to provide this assessment. This paper covers the first step in the analysis process, the development of a rapidly re-trainable, machine-learning, brain parcellation tool. The use of a sufficiently deep U-Net architecture encompassing a small (39 × 39 × 39 voxel input, 27 × 27 × 27 voxel output) sliding window to sample the entirety of the 3D image allows for the prediction of the entire image using only a single trained network. A large number of training, validating, and testing windows are thus generated from the 101 manually-labeled Mindboggle images, and full-image prediction is provided via a voxel-vote method using overlapping windows. Our method produces parcellated images that are highly consistent with standard atlas-based methods in under 3 min on a modern GPU, and the single network architecture allows for rapid retraining (<36 hr) as needed. [ABSTRACT FROM AUTHOR]

• This study proposes a method of automatic generation of architectural space based on user activities in a metaverse and analyzes users' perceptions of such spaces. • In reference to the theory of environmental psychology, we constructed a theoretical framework for quantitative analysis of activities, based on which, we developed a metaverse space. • Architectural space images were produced based on grammatical rules that converted the analysis results into text as well as a deep-learning-based text-to-image generator. • Twenty participants, after participating in the metaverse experience and receiving the architectural space image, responded to two questionnaires on their perceptions and opinions. • We analyzed the users' opinions on the architectural space created using the data they provided. Recently, architectural offices and IT companies have undertaken projects to build and sell cities in metaverses. However, it is considered inefficient to undertake such construction individually, building-by-building, using 3D modeling techniques. Fortunately, the use of artificial intelligence (AI) offers a breakthrough solution. This paper proposes a method of automatic generation of architectural spaces that is based on user activities in a metaverse and that can analyze users' perceptions of such spaces. To this end, in the present study, by referencing the theory of environmental psychology, we constructed a theoretical framework that can quantitatively analyze activities, based on which, we developed a metaverse space. Architectural space images were produced based on grammatical rules, which converted the analysis results into text, as well as a deep-learning-based text-to-image generator. Twenty participants responded to two questionnaires on their perceptions and opinions after participating in the metaverse experience and after receiving the architectural space image. Consequently, we analyzed the users' opinions on the architectural space created using the data they provided. Further, we reviewed the validity of the architectural design of the virtual space created using the proposed approach and explored the possibilities of its future utilization. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) has been proved to be an important engine of green economic development, yet how it will affect the internal structure of green economy is unknown. The aim of this study is to examine the impact and its mechanism of AI on green total factor productivity (GTFP) from the internal-structure perspective, by using provincial panel data of China from 2009 to 2021 and global Malmquist index. The main research results show that: (1) the development of AI contributes to China's GTFP growth. And this effect is more significant in undeveloped areas; (2) AI promotes China's GTFP growth mainly by improving resource allocation efficiency, while it exerts little impact through the paths of technological progress and scale efficiency; (3) the transmission mechanism of AI on GTFP varies greatly among China's three main regions. In the eastern region, AI improves GTFP mainly by both advancing technological progress and improving resource allocation efficiency, while in central region AI contributes to GTFP growth mainly through technological progress. Compared with the eastern and central regions, AI in the western region plays a stronger impact on GTFP through the channel of improving scale efficiency. This study helps to understand the pathways of artificial intelligence affecting the transformation of green economic growth and formulate differentiated regional policies in light of local conditions. • This paper studies the effects of AI on China's GTFP from internal-structural perspective. • GTFP is decomposed into three internal factors by the global Malmquist index. • AI has positive effect on GTFP through different impacts on internal factors. • Improving resource allocation efficiency is the most critical channel of AI on GTFP. • AI plays a more significant role in promoting GTFP in less developed areas. [ABSTRACT FROM AUTHOR]

Several cities in the developing world, of which the capital city of India, New Delhi, is an example, often experience air quality in which pollutant levels go way above the levels considered hazardous for human health. To bring down the air quality to within permissible limits quickly, the measures typically taken involve shutting down certain high-polluting activities for some time to enable the air quality to recover temporarily. This paper presents a first-ever model based on artificial neural networks to forecast the extent of reduction in air quality parameters that can be achieved and the time period within which a change can be experienced when the source of the emissions is cut off temporarily. The model is based on the extensive data on the extent of reduction in air quality parameters that occurred during the lockdown that was imposed during the COVID-19 pandemic. The non-linear autoregressive exogenous network-based model chosen for the purpose employs the hour since stopping of emissions, relative humidity, wind speed, wind direction, and ambient temperature as input parameters to predict the rate of change of PM 2.5 with respect to the concentration at the start of the stopping of the emissions. Air quality data from a key monitoring station in New Delhi was used to develop the model. The model predicted the rate of drop in PM 2.5 with an R and MSE of 0.0044 and 0.9736, respectively, while training and 0.0095 and 0.9583 while testing. The model was then tested with data from 19 other stations in New Delhi, and accuracy of the model was found to be exceptionally accurate, with the correlation between the measured and the predicted PM 2.5 levels ranging from 0.74 to 0.94 and the MSE ranging from 0.0110 to 1.0746. Thus, the model can be employed to determine the number of hours of temporary stoppage of emissions required for the PM 2.5 concentration to reach safe levels. The methodology of development of the model can be extrapolated to construct models tailored for use in other parts of the world as well. • A novel ANN -based model is reported to predict rate of drop of PM 2.5 • Model has been trained on extensive dataset from the covid lockdown in Delhi. • Duration and efficacy of emergency measures to reduce emissions can be estimated. • The model uses easy-to-measure meteorological parameters as inputs. • Model has been validated for 19 stations of various land-use types in Delhi. [ABSTRACT FROM AUTHOR]