Search

In scientific social networks, group information has become an important auxiliary information to enhance the performance of paper recommendation, as many researchers prefer to obtain interested papers by joining groups. However, the existing paper recommendation methods failed to make full use of group information. In this paper, a paper recommendation method considering group information with multi-graph attention fusion network (GI-MGAF) is proposed. Specifically, in the graph construction layer, we construct researcher-paper bipartite graph, group-researcher bipartite graph and group-paper bipartite graph. In the information propagation layer, graph attention networks (GAT) are used to learn the node representations on the constructed bipartite graphs. In the information fusion layer, the researcher-level attention and paper-level attention are developed to respectively fuse the representations of researchers and papers. Experiments were conducted on the real world CiteULike dataset and the results demonstrate the effectiveness of the proposed GI-MGAF method. [ABSTRACT FROM AUTHOR]

This paper investigates the effectiveness of sentiment analysis (SA) methods, ranging from rule-based approaches to deep learning architectures, in analysing textual data. The study focuses on three Python libraries: TextBlob, VADER, and Flair, evaluating their accuracy on a public dataset of Twitter posts. Additionally, custom neural network architectures are developed to optimize sentiment classification. Results indicate that while rule-based libraries offer simplicity, deep learning-based libraries show promise for higher accuracy. The customized LSTM models, particularly LSTM2 with architectural adjustments and regularization techniques, demonstrate improved performance over baseline models with classification accuracy as high as 76.3%. [ABSTRACT FROM AUTHOR]

This paper presents a critical review of the emerging field of eXplainable Artificial Intelligence (XAI) in machine condition monitoring (MCM), focusing on the integration of neural networks (NNs) for fault detection and diagnosis. As MCM increasingly adopts NNs for their accuracy in classification and prognostics, concerns about trust and safety in high-risk applications have risen, necessitating the development of XAI. XAI aims to interpret and justify neural network outputs, enhancing user trust in AI solutions. The study categorizes XAI approaches into two main families: feature importance analysis and structural XAI. Feature importance analysis, applied after network training, focuses on identifying input components used by neural networks in diagnostic conclusions. This approach builds trust by aligning network outputs with physically meaningful symptoms of machine faults. However, it is limited to explaining "what" the network considers, without shedding light on the "how" of its decision-making processes. Structural XAI, overlapping with the concept of physically-informed neural networks (PINNs), incorporates explainability into the network design stage. It involves structuring the network to reflect physically-justified mechanisms or tuning hyperparameters to mimic traditional signal-processing techniques. While offering insights into network functioning, this approach risks constraining the network's adaptability and innovative potential. The paper evaluates the potential and challenges of these XAI methodologies in MCM, particularly for tribological images and vibrational signals. It discusses the balance between ensuring network trustworthiness and maintaining the inherent flexibility and innovation of AI. Through detailed analysis, the paper underscores that while XAI provides valuable insights, overemphasis on explainability can undermine AI's unique strengths. It concludes that the integration of XAI in MCM should aim for an optimal balance, customizing the use of AI and analytical methods to specific monitoring needs, thus strengthening the understanding of the role of XAI in enhancing AI reliability and acceptance in condition monitoring. [ABSTRACT FROM AUTHOR]

This study is about sign language translation into text; this technology is for those experiencing difficulty in communicating, i.e., those who have speech disorders, hearing impairment, and are deaf. Sign language translation into text is done through a process when one uses sign language that could be translated into text. Deep learning is a machine learning and artificial intelligence (AI) technique that mimics how humans acquire knowledge; the complexity and abstraction of deep learning algorithms are stacked in a hierarchy. The researchers utilized Convolutional Neural Networks (CNN), Connectionist Temporal Classification (CTC), and Deep Belief Network (DBN) to collect and gather the results of their system through different methods of translating sign language into text. Through the systematic literature review of various published papers related to sign language translation using deep learning techniques, this paper discusses and systematically analyzed various studies and come up with a proposed model in translating sign language into text. [ABSTRACT FROM AUTHOR]

This paperproposes a new intelligent system for paper currency recognition. Pakistani paper currency has been considered, as a case study, for intelligent recognition. This paper identifies, introduces, and extracts robustfeatures fromPakistani banknotes. After extracting thesefeatures, the paper proposes to use three layers feed-forward Backpropagation Neural Network (BPN) for classification. The proposed technique and system are simple and comparatively less time consuming which makes it suitable for real-time applications. Implementation and experimentation of the proposed technique certify the authenticity to a high recognition. [ABSTRACT FROM AUTHOR]

The evaluation of musician performance has traditionally been subjective and challenging, relying on human judgment and personal preferences. However, recent advancements in artificial intelligence (AI) have opened up new possibilities for objective and automated assessment. This paper presents a comprehensive survey and analysis of AI-enabled techniques in evaluating musician performance. The research addresses the limitations of subjective evaluations and aims to enhance objectivity and efficiency in assessment. By leveraging AI and machine learning, various aspects of performance, including technical skills, musical interpretation, and emotional expression, can be objectively evaluated. The findings demonstrate the potential of AI-based assessment tools in providing objective feedback, enhancing music education, and enabling fair evaluations in competitions and auditions. The survey encompasses ten key papers that highlight different methodologies, advantages, and limitations in the field of automated music performance assessment. The reviewed literature showcases approaches such as deep neural networks, sensor localization algorithms, multimodal analysis, and machine learning techniques. Through a comparative analysis, this paper provides insights into the strengths and limitations of each approach. [ABSTRACT FROM AUTHOR]

To support market success, it is important to identify customer needs and the relation between customer needs and customer purchasing behavior. This paper provides an overview over existing, already established approaches to determine the relevance of customer benefits. Then, an approach utilizing artificial neural networks to correlate the attributes of battery electric vehicles and their sales performance is presented. This approach is discussed in relation to needs expressed by customers in surveys as well as typical user behavior of passenger cars. It seems that, for example, charging speed of electric vehicles is more important than driving range despite customers regularly expressing driving range as their greatest concern. The presented approach can be integrated into the reference process for developing product profiles and can be coupled with drive system optimization methods, to consider sales performance alongside vehicle performance, efficiency and costs in the early stage of product engineering. [ABSTRACT FROM AUTHOR]

Neural networks (NNs) have proven to be a very powerful tool both for one-dimensional (1D) and two-dimensional (2D) direction of arrival (DOA) estimation. By avoiding complex and time-consuming mathematical calculations, NNs estimate DOAs almost instantaneously. This feature makes them very convenient for real-time applications. Further, unlike the well known MUSIC algorithm, neural network-based models provide accurate directions without additional calibration procedure of antenna array and a priori knowledge of the number of sources. In this review paper, the results achieved by the research group at the Faculty of Electronic Engineering in Nis are presented. The problem of DOA estimation of narrowband signals impinging upon different configurations of antenna arrays is addressed. Both Multi-Layer Perceptron (MLP) and Radial Basis Function (RBF) neural networks are considered, and their advantages and disadvantages are discussed. To improve the resolution of DOA estimates, sectorization model is introduced. As shown in this work, neural network-based models demonstrate high-resolution localization capabilities and much better efficiency than the MUSIC. [ABSTRACT FROM PUBLISHER]

The relationship between online reviews and a firm declaring bankruptcy may seem unlikely. However, notwithstanding the lack of quality in relation to products or services, there is a strong anecdotal commentary on how firms lose business due to bad reviews, despite the perceived high quality of products. As such, this short paper attempts to evoke a critical dialogue on the question, "do instantiated or malicious review cause firms to lose business beyond recovery?" The paper provides exploratory evidence and attempts to provide a theoretical basis that explains this potential issue faced by millions of businesses around the world and contributes to identifying any possible pattern between negative online reviews and bankruptcy, theoretical foundations, and provides practical suggestions to mitigate the risk of bankruptcy for businesses by better managing the influx of negative reviews. Further, this paper will identify avenues for future research. [ABSTRACT FROM AUTHOR]

Sustainability assessment is a relevant issue nowadays and applies to various sectors, including healthcare. However, sustainability assessment requires dedicated tools and methods depending on the domain issue. Hence, this paper proposes a conceptual and structural framework for measuring the social sustainability of healthcare systems. The paper provides a new perspective on a complex assessment of the fundamental aspects of social sustainability with equity, quality, responsiveness, financial coverage, and adaptability in health systems. The approach that employs the multi-criteria PROMETHEE II method, based on the strong sustainability paradigm, assists health policy stakeholders in decision-making. The results demonstrate that healthcare systems in countries located in Northern and Western Europe, such as Norway, Germany, Netherlands, Belgium, Iceland, and Sweden, are the most socially sustainable. At the opposite end of the ranking are countries in Central and Eastern Europe: Poland, Latvia, Hungary, and Slovakia. [ABSTRACT FROM AUTHOR]

In this paper, we address the problem of pitch estimation using Self Supervised Learning (SSL). The SSL paradigm we use is equivariance to pitch transposition, which enables our model to accurately perform pitch estimation on monophonic audio after being trained only on a small unlabeled dataset. We use a lightweight (< 30k parameters) Siamese neural network that takes as inputs two different pitch-shifted versions of the same audio represented by its Constant-Q Transform. To prevent the model from collapsing in an encoder-only setting, we propose a novel class-based transposition-equivariant objective which captures pitch information. Furthermore, we design the architecture of our network to be transposition-preserving by introducing learnable Toeplitz matrices. We evaluate our model for the two tasks of singing voice and musical instrument pitch estimation and show that our model is able to generalize across tasks and datasets while being lightweight, hence remaining compatible with lowresource devices and suitable for real-time applications. In particular, our results surpass self-supervised baselines and narrow the performance gap between self-supervised and supervised methods for pitch estimation. [ABSTRACT FROM AUTHOR]

The prediction of disease can facilitate early intervention, comprehensive diagnosis and treatment, thereby benefiting healthcare and reducing medical costs. While single class and multi-class learning methods have been applied for disease prediction, they are inadequate in distinguishing between primary and secondary diagnoses, which is crucial for treatments. In this paper, label distribution is suggested to describe the diagnosis, which assigns the description degree to quantify the diagnosis. Additionally, a novel hierarchical label distribution learning (HLDL) model is proposed to make fine-grained predictions based on the hierarchical classification of diseases, taking into account the relationship among diseases. The experimental results on real-world datasets demonstrate that the HLDL model outperforms the baselines with statistical significance. [ABSTRACT FROM AUTHOR]

Having noisy data in datasets is not a rare situation. Noisiness in data can influence the results and accuracy of machine learning algorithms applications. This paper focuses on the question of which machine learning algorithm will perform the best when identifying noisy data. Answering this question brings us one step closer to a meta-learning recommendation system, the main goal of project SIMON. To answer this question, experiments are conducted on the publicly available datasets Bot-IoT, which consists of real and simulated IoT network traffic. Several approaches to implementing behavioral IoT botnet attack detection have been explored in the literature, including machine learning. In this paper, we are exploring four different types of machine learning algorithms: (i) machine learning algorithms based on error, (ii) machine learning algorithms based on information, (iii) machine learning algorithms based on similarity, and (iv) machine learning algorithms based on probability. The results show that the highest accuracy is achieved by a machine learning algorithm based on error - an artificial neural network that achieved the highest accuracy. However, even though the decision tree achieved slightly lower results, there is no statistically significant difference between the artificial neural network and the decision tree. [ABSTRACT FROM AUTHOR]

Face detection is one of the most researched fields in deep learning, but some aspects have not yet been sufficiently researched. In this paper, we will focus on the application of deep transfer learning for detecting face images with moustache. For the purpose of this paper, we use a version of the MobileNetV2 convolutional neural network which was pretrained on the ImageNet dataset and the provided weights are used for transfer learning to moustache detection problem. In addition, we created a balanced dataset for moustache detection from FFHQ dataset. Using our approach we were able to achieve a testing F1-score of 89%. [ABSTRACT FROM AUTHOR]

Determining the optimal trajectory of a manipulator is a mathematically complex part of control due to the time-consuming determination of inverse kinematics, the complexity of which increases as the manipulator's degrees of freedom increase. The control of manipulator actuators is based on algorithms focusing on the shortest possible control time and zero position deviation. Values such as power consumption, which affect the growth of the carbon footprint, are not taken into account. For this reason, alternative control methods are being investigated, some of which can be used to determine the optimal trajectory in an uncertain and variable environment paying attention to instantaneous power consumption. This paper examines the capabilities of an artificial neural network algorithm for determining the trajectory of a four-axis manipulator with special emphasis on power consumption. The chosen method for learning artificial neural networks was reinforcement learning. Unlike the classical approach to determining inverse kinematics, the solution presented in the paper is based on model output signals. The artificial neural network determines the control policy based on the angular position, angular velocities, power consumption of the manipulator's actuators, and feedback, which is the reward function. The network was trained to reach a desired point from a fixed and random initial state of the robotic arm in optimal time and effective power requirements in such a way as to maintain a set distance from periodically moving obstacles. A three-dimensional representation of the manipulator designed in a CAD environment and mathematical models of the manipulator's drives were used for training. This approach accelerated the learning process of the algorithm, and the steps taken guarantee very similar, if not identical, performance on a real object. [ABSTRACT FROM AUTHOR]

An analysis presented in the paper reveals that most equipment requires a DC power supply (low voltage and low current) rather than an AC power supply, like computers and peripherals, audio-video players, food preparation. Moreover, there is the possibility that through minimal adaptations and other equipment they can operate in DC microgrids. A validation for one or another microgrid types, when the AC grid has priority, is based on an in-depth study. There are currently consumers (especially in the home appliance field) that can be used successfully in bothmicrogrids (AC and DC). This category of receivers includes AC electrical light sources, which study and analysis of compatibility in DC network, will be detailed in the paper. The research results conduct to the conclusion that most of the AC consumers used nowadays, are easy to use and adapt to a DC microgrid, like the residential low voltage electric network. Paper will focus only to a detailed analysis for one lightning source. [ABSTRACT FROM AUTHOR]

Early identification and accurate diagnosis are crucial for successful therapy and improved patient outcomes of brain tumors. The study looks at how deep learning may help detect brain tumors, including how convolutional neural networks (CNNs) can change the face of diagnostic imaging. This underscores the need of precise and timely identification of brain cancer for efficient therapy and enhanced patient results. The study discusses the challenges associated with tumor segmentation and classification, such as variations in size, shape, and location. To overcome these challenges, deep learning techniques are utilized to analyze brain scans and identify tumor abnormalities. However, the paper underscores the importance of these models serving as tools to assist medical professionals rather than replacing their expertise. Additionally, the paper presents a comparative study that evaluates different techniques, artificial neural networks (ANN) achieve the highest accuracy at 99%, followed by Support Vector Machine (SVM) at 98.9%. vision transformer (ViT), Extreme Gradient Boosting (XG Boost), CNN-based dense Efficient Net, and Deep Neural Network (DNN) also show high accuracies above 98%. However, U-Net, Alex Net, hybrid ensemble method (Random Forest (RF), K-Nearest Neighbour (KNN), Decision Tree (DT), and Region-based convolutional neural networks (R-CNN)) have lower accuracies ranging from 91.66% to 97.305%. Overall, the findings highlight the potential of deep learning in enhancing the efficiency, precision of brain tumor diagnosis, leads to earlier treatment and enhanced patient prognosis. [ABSTRACT FROM AUTHOR]

The paper is focused on the determination of mechanical fracture parameters of alkali-activated concrete (AAC) specimens of nominal dimensions 80 × 80 × 240 mm with central notches of the following types: straight-through notch, chevron notch with 90° notch angle and chevron notch with 120° notch angle. In accordance with the results of the experimental research, a numerical simulation of the fracture test of the specimens tested in three-point bending was carried out. The 3D FEM computational model is based on the configuration of the laboratory test and was developed in ATENA-GiD software. Artificial neural network (ANN)-based inverse analysis was used to identify the mechanical fracture parameters of studied composite. A neural network ensemble was created, with one robust network for each type of notch, covering the entire domain of identified parameters. The main goal of this paper is to present the results of the identification of the mechanical fracture parameters of the AAC and their comparison with the values obtained by direct evaluation of the experiments. Another objective is to investigate the influence of the notch type on the mechanical fracture parameters of the AAC. [ABSTRACT FROM AUTHOR]

In this paper, we present the results of charged particle track reconstruction in CLAS12 using artificial intelligence. In our approach, we use neural networks working together to identify tracks based on the raw signals in the Drift Chambers. A Convolutional Auto-Encoder is used to de-noise raw data by removing the hits that do not satisfy the patterns for tracks, and second Multi-Layer Perceptron is used to identify tracks from combinations of clusters in the drift chambers. Our method increases the tracking efficiency by 50% for multi-particle final states already conducted experiments. The de-noising results indicate that future experiments can run at higher luminosity without degradation of the data quality. This in turn will lead to significant benefits for the CLAS12 physics program. [ABSTRACT FROM AUTHOR]

The Jiangmen Underground Neutrino Observatory (JUNO) experiment is set to begin data taking in 2024 with the aim of determining the neutrino mass ordering (NMO) to a significance of 3 σ after 6 years of data taking. Achieving this goal requires effective background suppression, with the background induced by cosmic-ray muons being one of the most significant sources of interference in the NMO study. Accurately simulating the cosmic-ray muon background is crucial for the success of the experiment, but the sheer number of optical photons produced by the muon makes this detector simulation process extremely time-consuming using traditional methods such as Geant4. This paper presents a fast muon simulation method that employs neural networks to expedite the simulation process. Our approach achieves an order-of-magnitude speed-up in simulation time compared to Geant4, while still producing accurate results. [ABSTRACT FROM AUTHOR]

The operating conditions defining the current data taking campaign at the Large Hadron Collider, known as Run 3, present unparalleled challenges for the real-time data acquisition workflow of the LHCb experiment at CERN. To address the anticipated surge in luminosity and consequent event rate, the LHCb experiment is transitioning to a fully software-based trigger system. This evolution necessitated innovations in hardware configurations, software paradigms, and algorithmic design. A significant advancement is the integration of monotonic Lipschitz Neural Networks into the LHCb trigger system. These deep learning models offer certified robustness against detector instabilities, and the ability to encode domain-specific inductive biases. Such properties are crucial for the inclusive heavy-flavour triggers and, most notably, for the topological triggers designed to inclusively select b-hadron candidates by exploiting the unique kinematic and decay topologies of beauty decays. This paper describes the recent progress in integrating Lipschitz Neural Networks into the topological triggers, highlighting the resulting enhanced sensitivity to highly displaced multi-body candidates produced within the LHCb acceptance. [ABSTRACT FROM AUTHOR]

There are ongoing efforts to investigate theories that aim to explain the current shortcomings of the Standard Model of particle physics. One such effort is the Long-Lived Particle Jet Tagging Algorithm, based on a DNN (Deep Neural Network), which is used to search for exotic new particles. This paper describes two novel optimisations in the design of this DNN, suitable for implementation on an FPGA-based accelerator. The first involves the adoption of cyclic random access memories and the reuse of multiply-accumulate operations. The second involves storing matrices distributed over many RAM memories with elements grouped by index. An evaluation of the proposed methods and hardware architectures is also included. The proposed optimisations can yield performance enhancements by more than an order of magnitude compared to software implementations. The innovations can also lead to smaller FPGA footprints and accordingly reduce power consumption, allowing for instance duplication of compute units to achieve increases in effective throughput. [ABSTRACT FROM AUTHOR]

The rise of cyber-attacks such as Distributed Denial of Service (DDoS) and SQL injection has become a significant concern for organizations and individuals who rely on the internet. Traditional detection methods have become increasingly ineffective in addressing these challenges, necessitating the development of new and innovative solutions. This paper proposes using Convolutional Neural Networks (CNNs) to detect DDoS and SQL injection attacks. Our paper proposes a Convolutional Neural Network model with a multi-layer neural and relu activation function optimizer that reaches a higher degree of precision than previous Deep Learning models. In this study, we tested the relatively new dataset CIC-IDS-2018, which contains different types of attacks. With this dataset, our model achieves an unprecedented accuracy of>96%, minimizing computational time. [ABSTRACT FROM AUTHOR]

The enormous benefits and applications of Image classification and recognition are umpteen. Machine learning algorithms and Deep Neural Networks are like windfall to fathom the objective proficiently in streamlined manner. The prevalent improvement in this technology is that these networks do not call for any prior blueprints in terms of algorithms as prerequisites. The presented paper is an attempt to create a Convolutional Neural Network from scratch to classify the images from the well-known dataset – Cats and Dogs into their relevant baskets. Manifold open source accessible approaches to amplify the efficiency of the network are no more onerous. Further, data augmentation technique boosts the efficiency tremendously by extending the dataset with reoriented features from the same images. To untangle the same problem, Transfer Learning is also a compelling technique in which all the layers, neurons in each layer, weights of each neuron and all other parameters are predefined and we can amend the output layer as per the classes in the respective problem statement. In the present paper, we have tried to obtain a comparable efficiency with a significant reduction in parameters. [ABSTRACT FROM AUTHOR]

This paper describes a case study of a single-family home to apply Six Sigma Problem Solving Methods and tools to reduce the electricity Bill and usage. The goal of the project was to identify and implement improvements that would reduce electricity usage and hence reducing electricity bill. The project team applied the Six Sigma DMAIC (Define, Measure, Analyze, Improve, Control) problem solving approach. Six Sigma is a framework to improve performance Process or Tasks. The author of the paper has been working on the topic over the past 1 year. The author was able to reduce the electricity bill by 50-60% by making some proper research and lifestyle change. The highest bill that was noted is $135 a month and lowest after performing the control measures came down to $57 a month. [ABSTRACT FROM AUTHOR]

In the last few years Artificial Intelligence (AI) algorithms and Machine Learning (ML) approaches have been successfully applied in real-world scenarios like commerce, industry and digital services, but they are not a widespread reality in Software Testing. Due to the complexity of software testing, most of the work of AI/ML applied to it is still academic. This paper briefly presents the state of the art in the field of software testing, applying ML approaches and AI algorithms. The progress analysis of the AI and ML methods used for this purpose during the last three years is based on the Scopus Elsevier, web of Science and Google Scholar databases. Algorithms used in software testing have been grouped by test types. The paper also tries to create relations between the main AI approaches and which type of tests they are applied to, in particular white-box, grey-box and black-box software testing types. We conclude that black-box testing is, by far, the preferred method of software testing, when AI is applied, and all three methods of ML (supervised, unsupervised and reinforcement) are commonly used in black-box testing being the “clustering” technique, Artificial Neural Networks and Genetic Algorithms applied to “fuzzing” and regression testing. [ABSTRACT FROM AUTHOR]

Nowadays, companies' first target is to build an intelligent, flexible, integrated and efficient processes, for higher industrial performance. According to this, a new concept emerged within the industrial firms by the name of Industry 4.0 (I4.0). This 4th industrial revolution covers four main aspects: intelligent factory, real-time communication, smart products and more satisfied customers. As strong as are the new technologies offered by this revolution, it is not entirely sufficient if not combined with lean principles. Combining these two paradigms created what is known today as Lean 4.0, a new concept presenting several opportunities for higher performance. Nevertheless, this combination is far from being as simple as it seems. A deep and structured risk analysis is required to ensure a smooth implementation of lean 4.0 system within a company. The purpose of this paper is to discuss I4.0, lean manufacturing and Lean 4.0 risks and opportunities through a systematic literature review, and to study the different barriers to its implementation through the modelling of a case study using sequence diagram and SWOT analysis. Its contribution consists of being a reflection paper to the companies aiming to develop their digital maturity by exploiting Lean 4.0 tools. [ABSTRACT FROM AUTHOR]

Revolution in medical imaging systems generate high-resolution images. This revolution has created new challenges in accessing, sharing, archiving, and protecting medical images. The number of these images has also increased significantly over the past decade, which is create another challenge. Recently, the health care industry has started noticing cloud computing as a solution because of its elasticity and scalability, which contribute to avoiding hardware obsolescence, providing universal access to medical images anywhere and anytime, and increasing their availability. This paper introduces the main characteristics of the cloud-based picture archiving and communication system (PACS) and highlights its technical and nontechnical advantages. Moreover, this paper illustrates challenges in cloud-based medical image exchange, regulation, and interoperability as well as how medical image existence can affect cloud users. Furthermore, the paper concludes with directions for future research in medical imaging. [ABSTRACT FROM AUTHOR]

The easy access to technology has led to mass production of information. This has created a challenge among many users in selecting or picking the relevant information often deeply buried in the text that is being massively produced. Automatic text summarization is fraught with many challenges that are well known from the ROUGE benchmarking results compared to human summarization. Further, the output of the summarization is oftentimes syntactically wrong. Since the process uses corpus, the semantic result of the output is also a challenging task. The challenge is to identify which technique can improve the quality of summarised texts. This paper present results of the summarization techniques used for abstractive text summarization, by looking at different studies based on techniques, ROUGE scores, and datasets used. The paper concludes by identifying the gaps where further research is required. [ABSTRACT FROM AUTHOR]

This paper describes an Artificial Neural Network (ANN) based speed estimation of BLDC drive using the three-phase current of the motor. The star point of three parallel resistors is used as the artificial star point of the motor. The comparators are connected to motor terminals and star point of resistors. This circuit determines the zero crossing of the back-EMF of the motor phases. Using zero crossing detection of the back-EMF with the estimated speed from the ANN, precise control of the BLDC motor is achieved, without the need for a sensor. The ANN is an offline trained network, trained using data obtained from simulations. The network is validated against different Inputs. The network is precise but at the same time simple enough to be implemented on a low-cost processor. The results are validated with simulations done in MATLAB using the trained neural network The findings and applications of neural networks to control the speed of PMSM are very well presented in the referenced papers, but the same for BLDC motors, especially those used in electric vehicles, have not been explored yet. The presented paper attempts to address the challenges and advantages of adopting this sensor-free approach for speed control in EVs. [ABSTRACT FROM AUTHOR]

Underestimation or overestimating activities' duration in construction projects leads to many problems like lack of funding or procrastination. However, estimating duration accurately for construction projects is difficult because of many factors like identified and unidentified negative risks. This paper tried to estimate realistic durations of steel framing activities in a high-rise building using a combination of Artificial Neural Networks (ANNs) and fuzzy logic algorithms with data from five similar projects. Initial planned durations and the top five pure risks are the inputs, and estimated actual durations are the outputs. The data from the first four projects was for training, and the fifth one was for simulation. Then, the estimated expected actual durations under uncertainties are compared to the actual durations of the project. The results showed that about 75% of the fifth project's activities had errors of less than three days compared to the actual durations. [ABSTRACT FROM AUTHOR]

Handwriting Detection is a computer's approach or ability to access and interpret understandable handwritten source inputs such as paper documents, touch screens, images, etc. Recognition of handwriting is characterized as the task of turning a language into its symbolic representation represented in its spatial form of graphical marks. The progress is base on artificial neural networks, which is an artificial intelligence area of study. To build a Handwriting character recognition system, similar techniques and methods are used. When the data is implemented and translated into digital form by an OCR (Optical Character Recognition) device, it can be edited, manipulated, and processed in a computer-understandable form. For extracting the characteristics of the handwritten alphabets, a method of feature extraction is implemented and then we use the data to train the artificial neural network. It contributes enormously to the development of an automation process and in various applications will enhance the interface between man and machine. [ABSTRACT FROM AUTHOR]

Agriculture is essential for everyone to promote sustainable development, the farming, which combines image processing, artificial intelligence, Deep learning, and Internet of Things (IOT). World population incensing every day. Due to the rising demand in the Agriculture industry, the need to collectively improve a plants and growth its field is very useful. In this paper, it is important to maintain the crop during its initial time, and also at period of harvesting. The image processing and artificial networks are used as a different techniques to maintain the detecting the diseases on the leaves and correct time to harvesting. When we take images with help of drones, the images are divided and changed to disease described three things vectors namely the first one is color, one more is texture and morphology. The vectors morphology gives 95% accuracy and its give more compare to other two vector features. This research paper proposed effective and useful algorithms for detection of disease with help implementation of Artificial Neural network algorithms using MATLAB. Detection of leaf or plant diseases with some manual techniques are requires a lot of work by maintaining a huge farm of crops, and it's very early stages it detects different types of symptoms to different diseases on plants, when the displayed on crop leaves. In this research paper survey on various disease classification techniques that can be for plant leaves diseases detection. For this purpose Artificial Intelligence, Neural network algorithms and back propagation techniques for adjustment of training data sets. [ABSTRACT FROM AUTHOR]

The wireless networks of the future generation are evolved as complex systems due to broadening in service prerequisites, application heterogeneity and networking of gadgets. In recent times, the major step forward of machine learning techniques is deep learning. However, in wireless/heterogeneous networks, the deep learning application for network traffic control is relatively new. The key disputes in the wireless backbone networks since the advancement of wireless networks are resource allocation and efficient network traffic control such as routing. In larger scale of networks and tangled radio environments, the method of adding intelligence to wireless networks is achieved by Deep Learning (DL). The tangled wireless networks supported with several nodes and their quality of variable link can be investigated by Deep Learning. This paper is presented with a perception of harnessing the next generation communication networks with the artificial neural networks. This work helps the readers to explore the unsolved issues to pursue their research and deeply understand the wireless network design with DL based state of the art facilities. In this work, we integrate the deep learning and wireless networking research with a widespread survey. Finally, this paper summarizes the disputes and benefits of acquiring ML and AI for next generation wireless systems. [ABSTRACT FROM AUTHOR]

Machine learning is constantly gaining ground in the mining industry. Machine learning-based systems take advantage of the computing power of personal, embedded and cloud systems of today to rapidly build models of real processes, something that would have been impossible or extremely time-consuming a couple of decades ago. The widespread access to the internet and the availability of cheap and powerful cloud computing systems led to the development and acceptance of tools to automate resource modelling processes or optimise mine scheduling, using machine learning methodologies. The domain modelling system discussed in this paper, called DomainMCF, has been developed by Maptek, using artificial neural network technology. In the application presented in this paper, DomainMCF is used to model the spatial distribution of marble quality categorical parameters, and the results are combined to produce a final marble quality classification using drillhole and quarry face samples from an operational marble quarry in NE Greece. DomainMCF was made available for this study as a cloud processing service through an early access program for individuals or companies interested in testing its capabilities and suitability in various modelling scenarios and geological settings. The resulting marble product classifications are compared with those produced by the already established classification system that is based on a more conventional estimation method. The produced results show that DomainMCF can be effectively applied to the modelling of marble quality spatial distribution and similar domaining problems. [ABSTRACT FROM AUTHOR]

This paper focuses on a comparison study between the traditional deep learning techniques like LSTMs and GRUs with the latest state-of-the-art transformers for the task of classifying a comment based on its toxicity. Specifically, four different deep learning models are built, implemented and trained on a standard dataset for comparing the performance of the models. The models that are explored and compared are Bi-directional LSTMs, GRUs, Bidirectional LSTMs with CNNs and Transformers (with pre-trained RoBERTa weights). [ABSTRACT FROM AUTHOR]

Considering the linear limitation of the traditional PID controller, a scheme combining the fuzzy system with BP neural network was proposed, that was, and the fuzzy neural PID controller. At the same time, in order to optimize the accuracy of parameter, this paper combined the chaos optimization with ant colony algorithm, namely, chaotic ant swarm (CAS). It, which took the error integral performance index as the objective function and the value range of design parameters and the minimum gain and phase margin as the constraint condition, established the optimization mathematical model, further optimized and trained the weight and bias error of the fuzzy BP neural network, and reduced the error caused by the fuzzy neural network used for function approximation. Comparing the traditional fuzzy PID controller and the PID controller based on chaos genetic algorithm (GA) with the PID controller designed in this paper, the simulation results show that the dynamic performance and steady-state accuracy of the system are improved well [ABSTRACT FROM AUTHOR]

Purpose -- Technological developments, emerging business trends, globalization and changes in tourist behaviour require new ways of measuring and forecasting business performance. The purpose of this paper is to provide a comprehensive insight into the importance of AI, primarily artificial neural networks (ANN), their usage and implementation in hotel industry. The crucial benefits of an implementation of ANN in hotel industry are reflected in their ability to collect, analyse and interpret enormous data quantities. On top of that, ANN is capable of an automated generation of advanced business models and forecasts. These capabilities enable a redirection of human effort to more meaningful activities. An improved in-depth quantitative analysis and digitalization provided by ANN significantly contributes to an improved business performance in hotel industry. Methodology -- An extensive desk-research of the existing literature was carried out. Secondary data was used to establish a conclusions concerning artificial neural networks usage potentials in hotel industry. Findings -- The application possibilities of ANN as a data analysis and forecasting tool that enhances business performance are practically endless. The growing importance of innovative technologies for a successful business practice makes scientific approaches and the necessary theoretical background inevitable. So far, ANN have provided better data analysis and forecasting results than any other quantitative forecasting method. This fact underscores the need for a collaboration between theoretical research and industrial practice. Contribution -- An extensive theoretical rationale is provided for the adoption of ANN as an innovative tool for data analysis and forecasting for hotel industry. ANN as a data analysis and forecasting tool redefines the issues of hotel industry performances. The combination of tourism and AI provides a new level of sophistication in hotel industry. [ABSTRACT FROM AUTHOR]

The energy sector is an exciting area of research and study due to the global demand for oil and gas, improving and optimizing related costs and risks, essentially maximizing the business output. Unplanned corrective maintenance can lead to unscheduled downtime if not attended effectively and efficiently. Maintenance can restore system healthiness and avoid catastrophic failures. This paper explores the prioritization of unplanned work orders (WO) for corrective maintenance using an Artificial Neural Network. A case study from Oil and Gas in Oman was investigated to check the actual practice. The paper proposes a new approach to prioritize unplanned maintenance work orders, considering a classification of three correlated features: failure severity, asset criticality, and reliability. The proposed method shows the needs of such correlated of these features. The Artificial Neural Network-based multi-layer perceptrons method is applied; 82.7% of work orders being tested from the case study shows the low probability of less than 50% on initial priority. The investigation reveals the effectiveness of the suggested method to be applied to get more priority insights. We recommend industrial practitioners to use the approach that supports prioritization of recourses and scheduling activities better, saving cost and avoiding system functional failures. [ABSTRACT FROM AUTHOR]

A major issue in urban settings is the heavy traffic and vehicle congestion due to poor traffic control or overpopulation. Nevertheless, this issue emphasizes the significance of proper traffic flow control and prediction to improve traffic management. This study provides a systematic literature review of novel twenty (20) papers that used the deep learning approach to solve the complex traffic management problem. The several studies gathered showed that long short-term memory (LSTM) models are effective for urban traffic flow prediction. The related studies shows that additional datasets, and a hybrid model, lead to better prediction accuracies. Based from the result of systematic literature, after criteria was se, this study proposed an AT-Conv-LSTM model that uses environmental data and data from social media. [ABSTRACT FROM AUTHOR]

Public spaces have played a huge role in the spatial transformation of South African cities over the past decades. On the other hand, gated communities have emerged due to the increase in the level of crime in cit ies in general, including those in South Africa. Their gro wth has been largely motivated by people's longing for safety and security, especially when in public spaces. This paper investigates the characteristics and functions of public spaces in gated communities, learning from the Irene Farm Villages Estate gated community. A mixed -method research design is adopted for this paper, which implies the use of both qualitative and quantitative methods to gather informat ion. This is an explanatory and descriptive research study, as it provides documentary information on how the concept of public space has been applied to the Irene Farm Villages Estate gated community. The results of this study reveal that further research is imperative to gain more users' perceptions regarding public spaces in gated communit ies. This paper recommends that there be more research conducted, specifically by the local government agencies, to enable a review of the long-term implications of gated communit ies, particularly on the integrated development aspect. [ABSTRACT FROM AUTHOR]

Tools and methodologies for distinguishing computer generated melodies from human-composed melodies have a broad range of applications from detecting copyright infringement through the evaluation of generative music systems to facilitating transparent and explainable AI. This paper reviews a data challenge on distinguishing computergenerated melodies from human-composed melodies held in association with the Conference on Sound and Music Technology (CSMT) in 2020. An investigation of the submitted systems and the results are presented first. Besides the structure of the proposed models, the paper investigates two important factors that were identified as contributors to good model performance: the specific music features and the music representation used. Through an analysis of the submissions, important melody-related music features have been identified. Encoding or representation of the music in the context of neural network modes are found noticeably impacting system performance through an experiment where the top-ranked system was re-implemented with different input representations for comparison purposes. Besides demonstrating the feasibility of developing an objective music composition evaluation system, the investigation presented in this paper also reveals some important limitations of current music composition systems opening opportunities for future work in the community. [ABSTRACT FROM AUTHOR]

Permanent magnet synchronous motor is used widely in industry, the performance requirements wouldn't be met by adopting traditional PID control in some of the occasions with high requirements. In this paper, a hybrid control strategy - nonlinear neural network PID and traditional PID parallel control are adopted. The high stability and reliability of traditional PID was combined with the strong adaptive ability and robustness of neural network. The permanent magnet synchronous motor will get better control performance when switch different working modes according to different controlled object conditions. As the results showed, the speed response adopting the composite control strategy in this paper was faster than the single control strategy. And in the case of sudden disturbance, the recovery time adopting the composite control strategy designed in this paper was shorter, the recovery ability and the robustness were stronger. [ABSTRACT FROM AUTHOR]

Developments in flight test data systems have led to an abundance of data which can be harnessed via machine learning for flight load estimations. Main measurements used for the load estimations are strains. However, the transition from strains to loads requires load equations obtained from calibration load tests. Estimated loads obtained by strain data are used for several purposes, from predictive maintenance to development projects. The fidelity of the obtained loads highly affects these activities. Prior research showed that higher accuracy could be achieved with artificial neural network models instead of classical regression analysis. However, the models provided some inaccurate results for some load components during the transition to predicting flight data, and further development is required to improve the flight loads predictions. Using the idea that calibration test data consists of the knowledge of the loads, our aim is to pass on this knowledge to the flight domain to increase the fidelity of the estimated loads. This is possible with machine learning algorithms classified under transfer learning. Transfer learning deals with the transfer of knowledge obtained in one domain to another domain. A load model developed using the calibration test data as a source for training artificial neural networks can be fine-tuned using flight test data, which also includes additional information to increase the accuracy. The research in this paper focuses on applying transfer learning to convey the knowledge in the calibration domain to the flight domain and shows the potential to overcome the poor generalisation in transition to flight in order to achieve high fidelity loads outputs when using such a method. [ABSTRACT FROM AUTHOR]

Driving while drowsy is one of the significant causes of road accidents, and detection of drowsiness of drivers is necessary to minimize such accidents. There are many methods proposed for drowsiness detection. In this paper, we present the details of a drowsiness detection system using deep learning techniques with ECG as its input. Continuous wavelet transforms (CWT) are applied on the input signal, segmented to 1-minute duration, and then it is converted into scalogram images to be fed as input to deep neural networks (DNN). We use pre-trained transfer learning models, AlexNet, ResNet-50, and VGG-19 for this work. The best results were obtained for Reset-50 with an accuracy of 88.31% without dropping the final layers, which is an improvement of 4.5% over the spectrogram-based implementation. This paper evaluates the effectiveness of wavelet-based features, and we note that the time-frequency domain features outperform. To check the robustness of the net-work, a second model is implemented by dropping the final layers, and the validation accuracy of AlexNet is 51.5%, ResNet-50 is 67.2%, and VGG-19 is 81.8%. [ABSTRACT FROM AUTHOR]

This paper presents a proposed design and investigation of an impressed current and Single-Ended Primary Inductive Converter (SEPIC) converter based on an artificial neural network to assess the potential difference required to control cathodic protection system. The cathodic protection methods proved their effectiveness in the metallic equipment protection in the industrial applications that contain liquids and that are buried in electrolyte mediums against the electrochemical corrosion process, where D.C power supply plays an important role in the cathodic protection system. The proposed design consists of an intelligent d.c power supply and a group of anodes. This model has been adjusted to be able to equalize the normal voltage difference between the surrounding medium and the metal part being protected. The proposed design has been tested and investigated the cooling water carrying pipeline of AL-Quds electrical power station as a part of the unified Iraqi power network. The protection voltages and currents were calculated with a permissible tolerance based on; the pipeline characteristics, its coating, and the buried soil along the pipeline length to ensure better performance of protection. The obtained experimental and simulation results showed high effectiveness and accuracy by maintains the negative potential of the pipeline within the permissible and standard limits of the Iron metal of the pipeline. [ABSTRACT FROM AUTHOR]

Weather prediction is a key aspect of today's society and its activities. Accurate predictions are crucial for efficiently organizing human activities, particularly in the agricultural, transportation and energy sectors. In this paper, two deep neural networks, designed based on a long short-term memory architecture, are developed to predict the occurrence of rainfall events and the respective amount of rainfall on the island of Nisyros in the south Aegean Sea. Two deep neural networks are developed, serving two different learning tasks. The first network acts as a classifier that assesses whether it is going to rain or not and, sequentially, the second network performs a regression task, quantifying the anticipated amount of rainfall. The performance of such prediction models is highly dependent on input sequences. Among others, the lookback time window shapes those input sequences by determining the number of past time steps to be taken into account. The ideal time window for the classifier involves 24 time steps (i.e., 4 h), resulting in increased accuracy levels exceeding 96.45%. The predictions of the regression model, which has the same lookback time window, feature low errors, measured as 6.635 and 1.411 mm using the mean-square-error and mean-absolute-error indices, respectively. [ABSTRACT FROM AUTHOR]

Stack Overflow is a popular Q&A platform for developers to find solutions to programming problems. However, due to the varying quality of user-generated answers, there is a need for ways to help users find high-quality answers. While Stack Overflow's community-based approach can be effective, important technical aspects of the answer need to be captured, and users' comments might contain doubts regarding these aspects. In this paper, we showed the feasibility of using a machine learning model to identify doubts and conducted data analysis. We found that highly reputed users tend to raise more doubts; most answers have doubt in the first comment, and many answers have unsolved doubt in the last comment; high-score and low-score answers are equally likely to contain doubts in comments. Our classifier and findings can provide users with a new perspective on determining answers' helpfulness and allow expert users to easily locate doubts to address. [ABSTRACT FROM AUTHOR]

The rapid growth of social media platforms has led to an easier circulation of information around the world. There are a growing number of articles proposing various interventions to attempt to curb the spread of misinformation. This systematic literature review investigates the published research on misinformation interventions on social media platforms and provides an overview of existing literature on this subject by analyzing 346 papers included in major academic databases. We identify three categories of research, namely behavioral, modeling, and technical research, and classify various approaches proposed in behavioral research into three groups based on the research perspective, namely information, users, and official channels. Recommendations and guidance for future research are finally provided. [ABSTRACT FROM AUTHOR]

This paper proposes a new music source separation (MSS) model based on an architecture with MLP-Mixer that leverages multilayer perceptrons (MLPs). Most of the recent MSS techniques are based on architectures with CNNs, RNNs, and attention-based transformers that take waveforms or complex spectrograms or both as inputs. For the growth of the research field, we believe it is important to study not only the current established methodologies but also diverse perspectives. Therefore, since the MLP-Mixer-based architecture has been reported to perform as well as or better than architectures with CNNs and transformers in the computer vision field despite the MLP's simple computation, we report a way to effectively apply such an architecture to MSS as a reusable insight. In this paper we propose a model called TFC-MLP, which is a variant of the MLP-Mixer architecture that preserves time-frequency positional relationships and mixes time, frequency, and channel dimensions separately, using complex spectrograms as input. The TFC-MLP was evaluated with source-to-distortion ratio (SDR) using the MUSDB18-HQ dataset. Experimental results showed that the proposed model can achieve competitive SDRs when compared with state-of-the-art MSS models. [ABSTRACT FROM AUTHOR]