Neural Network-based System Research Articles

Seagull-optimized artificial neural network-based innovative energy management system using improvised boost converter and coupled inductor

Seagull-optimized artificial neural network-based innovative energy management system using improvised boost converter and coupled inductor

The Mason-Alberta Phonetic Segmenter: a forced alignment system based on deep neural networks and interpolation.

Given an orthographic transcription, forced alignment systems automatically determine boundaries between segments in speech, facilitating the use of large corpora. In the present paper, we introduce a neural network-based forced alignment system, the Mason-Alberta Phonetic Segmenter (MAPS). MAPS serves as a testbed for two possible improvements we pursue for forced alignment systems. The first is treating the acoustic model as a tagger, rather than a classifier, motivated by the common understanding that segments are not truly discrete and often overlap. The second is an interpolation technique to allow more precise boundaries than the typical 10 ms limit in modern systems. During testing, all system configurations we trained significantly outperformed the state-of-the-art Montreal Forced Aligner in the 10 ms boundary placement tolerance threshold. The greatest difference achieved was a 28.13 % relative performance increase. The Montreal Forced Aligner began to slightly outperform our models at around a 30 ms tolerance. We also reflect on the training process for acoustic modeling in forced alignment, highlighting how the output targets for these models do not match phoneticians' conception of similarity between phones and that reconciling this tension may require rethinking the task and output targets or how speech itself should be segmented.

Deep Neural Network-Based Cigarette Filter Defect Detection System with FPGA Acceleration for Online Recognition

In the cigarette manufacturing industry, machine vision and artificial intelligence algorithms have been employed to improve production efficiency by detecting product defects. However, achieving both high accuracy and real-time defect detection for cigarettes with complex patterns remains a challenge. To address these issues, this study proposes a model based on RESNET18, combined with a feature enhancement algorithm, to improve detection accuracy. Additionally, a method is designed to deploy the model on a field-programmable gate array (FPGA) with high parallel processing capabilities to achieve high-speed detection. Experimental results demonstrate that the proposed detection model achieves a detection accuracy of 95.88% on a cigarette filter defect dataset with an end-to-end detection speed of only 9.38 ms.

Research on Intelligent Agricultural Monitoring and Control System Based on Neural Networks in the Internet of Things Environment

In today's rapidly advancing information technology landscape, the Internet of Things (IoT) has profoundly impacted traditional agriculture. IoT technology, as an intelligent solution, enables real-time data collection, processing, and analysis, significantly enhancing the efficiency and intelligence of agricultural production. Among these advancements, the neural network-based intelligent agricultural monitoring system stands out. This system not only monitors the agricultural production environment in real-time but also employs deep learning algorithms to predict future trends, guiding agricultural practices to achieve precision farming. Therefore, exploring the intelligent agricultural monitoring and control system based on neural networks in the IoT environment holds significant importance.

CNN-Based Voice Recognition by Self-Governing Robots to Improve Computer-Human Communication

Robots must have the ability to recognize human emotions to engage with individuals and to plan their movements autonomously. Nonverbal signals encompassing pitch, loudness, spectrum, and speech speed are successful methods for transmitting emotions to most individuals. Provided this situation, a machine could have the capability of qualified figure out emotions by deploying the traits of spoken communication, and these potentially propagate vital information concerning the speaker's emotional state. More precisely, a combination of numerous facial action units can be employed to describe a human's emotion. In this paper, we propose a deep Convolutional Neural Network-based system that can identify sentiments in real-time and with a high accuracy rate. This investigation establishes an entirely novel speech-emotion detection system founded on Convolutional Neural Networks (CNNs). With the support of a top-tier GPU, a model is developed and nourished raw speech from a specific set for training, classification, and testing purposes. We further analyze the speech data and incorporate the information from the visual and audio sources to enhance the recognition system's accuracy. The benefits of the proposed method for emotion identification and the implications of blending visual and aural suggestions are made clear by the experimental results. Convolutional neural networks (CNNs) are taught on grayscale images from the softmax dataset in the present work. To acquire the best accuracy, we experimented with different baselines and max pooling layers, finally acquiring 89.98% accuracy. Dropout is one approach we have used to ward off overfitting.

Design of Neural Network-Based Multi-Fault Tolerant Control System for Unmanned Aerial Vehicles

Actuator failure seriously threatens the flight safety of unmanned helicopters. Considering the problem of multiple faults such as actuator bias and failure in unmanned helicopters, a composite fault-tolerant flight control algorithm is proposed. For actuator bias fault, a nonlinear fault observer is designed to estimate it in real time; for actuator failure fault, a same-dimensional auxiliary system is constructed and processed by neural network technology. The composite fault-tolerant flight controller of the unmanned helicopter is designed by backstepping method, and the Lyapunov stability theory is used to prove that the error signals of the closed-loop system are bounded and convergent. Simulation results show that the proposed control algorithm can improve the fault tolerance of the unmanned helicopter when multiple actuator faults occur, ensuring its safe flight.

Analytical Neural Network System for the Helicopter Turboshaft Engines Operating Modes Classification

A novel neural network-based analytical system has been developed for classifying helicopter turboshaft engine operating modes during flight operations. This system utilizes a neural network architecture comprising an input layer with three neurons, two fully connected hidden layers, and an output layer with two neurons. The proposed approach demonstrates an exceptional recognition accuracy of 0.997 (99.7%) across steady-state, unsteady, and transient operating modes of helicopter turboshaft engines. A new method for training this neural network has been introduced, employing forward propagation, loss calculation, backpropagation, and weight updates, enhanced by an adaptive learning rate and the cross-entropy function as the loss criterion. The method also incorporates a novel modified Smooth ReLU activation function for hidden layer neurons. This innovation led to a near-perfect accuracy in network training and reduced the loss to 0.025 (2.5%), highlighting the high quality and reliability of the neural network in classifying engine operating modes during flight. Furthermore, it has been empirically shown that the application of this neural network significantly reduces type I errors by a factor of 2.09 to 2.14 and type II errors by 2.05 to 2.21 times compared to traditional classifiers based on ART-1 and BAM networks. This advancement marks a substantial improvement in classification accuracy and error minimization for helicopter turboshaft engine operating modes.

PlugGuard: A Neural Network-Based Power Quality Control System for Plug Loads

PlugGuard: A Neural Network-Based Power Quality Control System for Plug Loads

A Convolutional Neural Network-based Intelligence System for the Identification of Copra Maturity Levels

The North Sulawesi Province, widely recognized as the Coconut Waving Province owing to its substantial coconut tree population, primarily depends on copra production. This research presents a novel methodology for determining copra maturity levels by utilizing a Convolutional Neural Network (CNN) on digital photographs, classifying them into three distinct stages: raw, half-ripe, and ripe. By employing a rigorous 10-fold cross-validation technique, our models demonstrated remarkable performance. Notably, even the model with the lowest performance achieved a commendable accuracy of 87.78% during the training and validation phases. The model that exhibited the highest level of performance achieved a perfect accuracy rate of 100%. Moreover, when subjected to real-world testing situations using novel data, the model with the lowest performance exhibited a noteworthy accuracy of 83.34%. In contrast, the highest-performing model achieved a flawless accuracy of 100%. Based on the findings above, an online system has been built that leverages the most optimal model, facilitating the assessment of copra maturity in real-time. The prospects encompass the integration of this methodology into copra sorting machinery, thereby yielding advantages for both agricultural producers and industrial sectors. This research enhances copra quality control processes and promotes sustainability in the copra industry. Further research could explore refining the CNN model to accommodate a broader range of copra variations and investigating automation possibilities in copra production processes. These endeavors would advance the efficacy and applicability of copra maturity classification methods, fostering continued innovation in the industry.

Predicting storm surge extremes on the Southeast Brazilian Coast: Long-term projections with neural networks

Extreme sea level events, resulting from the confluence of tides and storm surges, pose a significant threat to coastal populations and economies. The escalating risks associated with these events are exacerbated by climate change, manifesting in heightened storm intensity, increased frequency, and rising sea levels. Precise estimation of the probability of extreme storm surges is crucial for effective coastal management and adaptation. However, utilizing historical storm data is challenging due to data scarcity and the imperative to consider potential non-stationarity induced by climate change in predicting such events. This study addresses these challenges by introducing two neural network-based machine learning systems: a Multilayer Perceptron (MLP) and a Long Short-Term Memory (LSTM). Leveraging local and remote atmospheric and oceanic conditions, these systems project storm surges until 2060, incorporating climate projections. Trained and evaluated using sea level data from Imbetiba Port in Macaé, Rio de Janeiro, Brazil, the models utilize dynamic regionalization data from the RegCM4 and WW3 models, forced by HadGEM2-ES and MPI climate models. Both neural network models exhibited similar performance patterns, demonstrating high agreement in predicting storm surge heights with a 100-year return value, based on Imbetiba Port data. Projections utilized peaks-over-threshold (POT) methods, and extremes were calculated using a generalized Pareto distribution (GPD). Long-term projections indicated a 28 % increase (MLP ANN) and a substantial 70 % increase (LSTM RNN) in estimating extreme values, surpassing the observed storm surge of 0.67 m. Projected mean values were 0.86 m for the MLP network and 1.15 m for the LSTM network, providing valuable insights into the potential amplification of extreme sea level events in the studied region.

Hypergraph contrastive learning for recommendation with side information

PurposeThis paper addresses the limitations of current graph neural network-based recommendation systems, which often neglect the integration of side information and the modeling of complex high-order interactions among nodes. The research motivation stems from the need to enhance recommendation performance by effectively utilizing all available data. We propose a novel method called MSHCN, which leverages hypergraph neural networks to integrate side information and model complex interactions, thereby improving user and item representations.Design/methodology/approachThe MSHCN method employs a hypergraph structure to incorporate various types of side information, including social relationships among users and item attributes, which are essential for enriching user and item representations. The k-means clustering algorithm is utilized to create item-associated hypergraphs, while sentiment analysis on user reviews refines the modeling of user interests. Additionally, hypergraphs are constructed for user-user and item-item interactions based on interaction similarity. MSHCN also incorporates contrastive learning as an auxiliary task to enhance the representation learning process.FindingsExtensive experiments demonstrate that MSHCN significantly outperforms existing recommendation models, particularly in its ability to capture and utilize side information and high-order interactions. This results in superior user and item representations and improved recommendation performance.Originality/valueThe novelty of MSHCN lies in its use of a hypergraph structure to integrate diverse side information and model intricate high-order interactions. The incorporation of contrastive learning as an auxiliary task sets it apart from other hypergraph-based models, providing a significant enhancement in recommendation accuracy.

A Visual Transformer and Convolution Neural Network-Based Intelligent Recommender System for e-Commerce Scenes

Recommender systems have been a kind of powerful tool to improve e-commerce benefits. Existing recommender systems mostly employ explicit features of products, such as description and attributes. However, visual characteristics contain fruitful implicit and intuitive information, and are always ignored by existing works. To deal with this issue, this paper proposes a novel intelligent recommender system via visual Transformer (ViT) model and convolutional neural network (CNN) structure. First, the ViT part is utilized to extract visual feature representation. By learning the visual similarity among products, it is expected to obtain a better scene understanding. Then, an improved CNN part is utilized to extract hidden association information from historical behaviors of users. It is expected to better perceive user preference and purchasing characteristics. The combination of two parts constructs the proposed recommender system. Finally, we make performance evaluation for the proposal on real-world e-commerce dataset. The results indicate that the proposal exhibits high recommendation accuracy and efficiency. Compared with other typical algorithms, our proposal can better understand product images and user behaviors, providing more personalized recommendation results.

Environmentally Sustainable Detection of Arsenic using Convolutional Neural Networks and Imidazole-Based Organic Probes: Application in Food Samples and Arsenic Album.

Arsenic contamination poses a significant health risk, particularly when it infiltrates water supplies. While current detection methods offer precise analysis, they often involve complex instrumentation not suitable for field use. This study presents a novel approach by developing two probes, A1 and A2, based on 4-diethylaminosalicyladehyde, 2-hydroxy-1-naphthaldehyde, and 1,2-diaminoanthraquinone. These probes are highly sensitive and selective for detecting arsenite (As(III)) and arsenate (As(V)) in water, food samples, and homeopathic medicine with limits of detection in the nanomolar range. To elaborate our contribution to on-site arsenic detection, we introduce a convolutional neural network-based image recognition system. This system interprets images of the probes' colorimetric response, effectively categorizing different ranges of arsenic concentrations in parts per million (ppm). Our approach offers a real-time, cost-effective, and user-friendly solution for arsenic detection, extending its applicability from scientific laboratories to in-field conditions and even household monitoring. The findings fill critical research gaps in real-time detection methods, potentially revolutionizing the way we monitor environmental contaminants like arsenic.

Deep Feed-Forward Neural Network-Based Biometric Authentication System with Biometric Identity and Reputation Score in Blockchain

The biometric authentication in today’s digital era has become a vital issue in security and privacy. Biometrics is currently becoming more significant specifically for higher security modules. Here, a deep feed-forward neural network-based biometric authentication system (DFFNN_biometric authentication system) is presented for biometric authentication utilizing a biometric fingerprint image. In blockchain network, biometric data are considered and the fingerprint images are fed as input. As an input fingerprint image, minutiae extraction is accomplished. Thereafter, a deep key is generated employing a deep residual network (DRN) for template protection. Thereafter, privacy protection template generation is conducted and the templates thus obtained are stored on a database (DB). On the other hand, a query fingerprint image is given as input in the authentication stage. The processes such as minutiae extraction, deep key generation using DRN and privacy protection template authentication are carried out as like it performed in the blockchain network. Additionally, the matching process is conducted between the output acquired in the authentication phase and image stored in DB. The reputation score is generated from authenticated biometric behavior. The DFFNN is utilized to generate a reputation score. Moreover, the DFFNN_biometric authentication system obtained a maximum accuracy of 90.8%, maximum genuine accept rate (GAR) of 0.936, minimum false acceptance rate (FAR) of 0.588, minimum false rejection rate (FRR) of 0.556 and maximum reputation score of 9.346.

Reliable prediction of peak cladding temperature during top reflooding processes: A neural network-based perception-interaction approach

The reliable prediction of peek cladding temperature is crucial for heat transfer and reactor security when large break loss of coolant accident (LBLOCA) happens. This study proposes a neural network-based Perception-Interaction system to effectively predict and monitor the peak cladding temperature during reflooding phase after LOCA. The system employs a narrow rectangular channel for conducting top reflooding experiments, wherein the peak temperature is predicted by the Back Propagation Neural Network (BPNN) algorithm. The system’s primary objective is to accurately predict the peak cladding temperature of various test conditions. First, the special Counter-Current Flow Limitation (CCFL) phenomenon is identified based on the different heat exchange patterns inside the test section. Then the experimental data is collected by LabVIEW and calculated by MATLAB to accurately predict the quench velocity and peak temperature at a specific location by conducting various test conditions. The final experiment results indicate that the built Perception-Interaction system in this article has a high prediction accuracy, with a relative error within 2% between the predicted peak temperatures and actual temperatures.

A Neural Network-Based Scoring System for Predicting Prognosis and Therapy in Breast Cancer.

Breast cancer is a prevalent malignancy affecting women worldwide. Currently, there are no precise molecular biomarkers with immense potential for accurately predicting breast cancer development, which limits clinical management options. Recent evidence has highlighted the importance of metastatic and tumor-infiltrating immune cells in modulating the antitumor therapy response. However, the prognostic value of using these features in combination, and their potential for guiding individualized treatment for breast cancer, remains vague. To address this challenge, we recently developed the metastatic and immunogenomic risk score (MIRS), a comprehensive and user-friendly scoring system that leverages advanced bioinformatics methods to facilitate transcriptomics data analysis. To help users become familiar with the MIRS tool and apply it effectively in analyzing new breast cancer datasets, we describe detailed protocols that require no advanced programming skills. © 2024 Wiley Periodicals LLC. Basic Protocol 1: Calculating a MIRS score from transcriptomics data Basic Protocol 2: Predicting clinical outcomes from MIRS scores Basic Protocol 3: Evaluating treatment responses and guiding therapeutic strategies in breast cancer patients Basic Protocol 4: Guidelines for utilizing the MIRS webtool.

Retraction Note: Artificial neural networks (ANN), MARS, and adaptive network-based fuzzy inference system (ANFIS) to predict the stress at the failure of concrete with waste steel slag coarse aggregate replacement

Retraction Note: Artificial neural networks (ANN), MARS, and adaptive network-based fuzzy inference system (ANFIS) to predict the stress at the failure of concrete with waste steel slag coarse aggregate replacement

IoT-based trusted wireless communication framework by machine learning approach

The traditional Radio-Frequency Systems (RFS) authentication methods, designed to ensure secure data transmission on the web, may not always effectively prevent adversaries from gaining access to concealed IDs or asymmetric cryptography through infiltrative, side-channel, training, and computer attacks. In contrast, Unaccounted Information (UAI) has the potential to exploit irregularities in production systems to automatically identify microchips, offering a highly robust and cost-effective security solution. This approach introduces RFS-UAI, a deep neural network-based system that efficiently manages wireless node identification by leveraging synthetic RFS characteristics of remote controls (Tx) learned through supervised methods in Wireless Sensor Networks (WSN). Unlike traditional methods that require the development of specialized transistors for UAI or semantic segmentation, this approach utilizes the existing asymmetrical RFS communication networks. Similar to the way the human brain processes information, Rx handles the entire device identification process at the gateway. According to test results, which include assessing process capability at a specified 65 nm threshold voltage and characteristics such as Local Oscillator (LO) misalignment and I-Q disparity using a probabilistic model with 52 hidden units, the system can distinguish up to 4800 transmitters with a remarkable 99.9 % accuracy under various channel conditions, all without the need for regular preambles. This recommended method can serve as a standalone security measure or be integrated into a biometric identification system.

Deep Graph Neural Networks for Malware Detection Using Ghidra P-Code

This work examines the effectiveness of using Ghidra P-Code as semantics-based features in a graph neural network-based malware detection system. A preliminary model exhibits a function level precision of ∼70% and a recall around ∼60%, and a precision and recall of ~55% and ~80% respectively for the program level detection task on a dataset of ∼50,000 control flow graphs extracted from functions of malicious and benign programs. Future improvements to this ongoing project include, but are not limited to, collecting dynamic control flow graph information as opposed to static graphs to provide the model with resilience to advanced malware obfuscation and encryption schemes.

IoT-Enabled Single-Camera Speed Sensor for Smart City Tasks

This article presents an innovative IoT-enabled single-camera speed sensor designed for smart city applications. The research encompasses the development of both hardware and software components, focusing on a computer vision and artificial neural network-based system. A novel aspect of this study is the implementation of a distance-measuring algorithm that eliminates the need for expensive LIDAR sensors traditionally used in speed cameras. Instead, the system relies on convolutional neural networks (CNN) and computer vision algorithms to estimate vehicle speed accurately. Field testing was conducted in real conditions over several months, generating sufficient data to assess the device’s ability to function under various adverse conditions. The results demonstrate the system’s capability to perform vehicle speed detection consistently across diverse conditions, showcasing its potential as a scalable solution for urban traffic management. This makes the proposed system not only more affordable but also simpler to deploy and maintain, thereby enhancing its suitability for widespread use in smart city environments.