Network Error Research Articles

Near-infrared quality monitoring modeling with multi-scale CNN and temperature adaptive correction

Considering the impact of temperature on near-infrared spectroscopy (NIRS), a temperature adaptive correction neural network model called TI-CNN has been developed. This model takes into account the temperature influence. First, a multi-scale CNN model is established to extract the global features of NIRS. Then, by establishing the quantitative relationship between the sample and the modeling error of the neural network, the influence of temperature on the spectral features. Based on the degree of influence, the weight is adaptively adjusted to quantify the effect of temperature on near-infrared spectral features. Finally, to verify the model performance, EPO-PLSR, OSC-PLSR, CNN, and TA-CNN were used for comparison on the test set. The results showed that traditional temperature correction models performed poorly, with a coefficient of determination R2< 0.80. However, the neural network models based on multi-scale CNN, R2> 0.80. Of these, the TI-CNN based on multi-scale CNN and temperature adaptive correction, demonstrated the best performance with R2> 0.95. Therefore, the method proposed in this paper demonstrates superior performance compared to existing methods. It utilizes a multi-scale CNN to extract near-infrared spectral features and also incorporates temperature to adaptively correct the spectral features, resulting in superior model performance across various temperature conditions.

Deep Optimized Broad Learning System for Applications in Tabular Data Recognition.

The broad learning system (BLS) is a versatile and effective tool for analyzing tabular data. However, the rapid expansion of big data has resulted in an overwhelming amount of tabular data, necessitating the development of specialized tools for effective management and analysis. This article introduces an optimized BLS (OBLS) specifically tailored for big data analysis. In addition, a deep-optimized BLS (DOBLS) network is developed further to enhance the performance and efficiency of the OBLS. The main contributions of this article are: 1) by retracing the network's error from the output space to the latent space, the OBLS adjusts parameters in the feature and enhancement node layers. This process aims to achieve more resilient representations, resulting in improved performance; 2) the DOBLS is a multilayered structure consisting of multiple OBLSs, wherein each OBLS connects to the input and output layers, enabling direct data propagation. This design helps reduce information loss between layers, ensuring an efficient flow of information throughout the network; and 3) the proposed methods demonstrate robustness across various applications, including multiview feature embedding, one-class classification (OCC), camera model identification, electroencephalogram (EEG) signal processing, and radar signal analysis. Experimental results validate the effectiveness of the proposed models. To ensure reproducibility, the source code is available at https://github.com/1027051515/OBLS_DOBLS.

Real-time prediction of wave-induced hull girder loads for a large container ship based on the recurrent neural network model and error correction strategy

Real-time acquisition of wave-induced hull girder loads of a sailing ship will help the captain make reasonable decisions, which is of great significance for improving the safety of the ship's navigation. This paper investigates the real-time prediction method of hull girder loads based on the Recurrent Neural Network (RNN) model and error correction strategy. Firstly, taking the vertical bending moment, horizontal bending moment, and torsional moment at the mid-ship position of a large container ship as examples, corresponding neural network prediction models are established through parameter influence analysis. Secondly, various sea state conditions are used to verify the feasibility of established network prediction models to predict the hull girder loads in real-time. The VBM prediction model performs better than the TM prediction model and HBM prediction model, and the errors of the TM prediction model and HBM prediction model are slightly larger in some cases. Lastly, an improved prediction model based on an error correction strategy is proposed to improve the prediction accuracy of the neural network prediction model, and the adequate performance of the error correction strategy is discussed.

Evaluation of Blended Learning in Terms of the Perceptions of Medical Students: A Mixed Method Study.

The ongoing 4.0 industrial evolution, characterized by the rise of digital technology, has had a massive impact on human lifestyles worldwide. Faculty members in medical school are expected to respond to this industrial revolution by implementing teaching strategies, one of which is Blended learning as a suitable solution to overcome the limitations of space and time in the teaching process. For effective utilization of blended learning, it is important to conduct extensive studies on its implementation. The aim of this study was to assess the effectiveness and efficiency of implementing blended learning in the faculty of medicine in Hasanuddin University from the students' perspective. This study used a sequential explanatory mixed method approach, combining quantitative and qualitative methods. The quantitative part involved 782 undergraduate medical students from the first, second, and third years. Data were collected through a questionnaire survey distributed among the students. The qualitative part of the research was conducted through focus group discussions involving 13 students based on the questionnaire scores, representing both high and low scores. The results of the quantitative and qualitative research were collected and integrated. Based on the results, the majority of students agreed that blended learning provided many advantages to their learning (Mean±SD: 3.79±0.78). Also, they reported e-learning platform significantly contributed to their learning process (Mean±SD: 3.88±0.67). The workload of blended learning method was still considered quite heavy by students, and good time management was highly needed (Mean±SD: 3.45±0.84). As for qualitative part, some positive results were obtained; they reported that it increased motivation for learning, enhanced the efficiency of learning and gaining adaptability, while the negative opinions were the network error in e-learning, erratic e-learning display, and video quality problem. Most of the students expressed positive opinions about the advantages of blended learning; according to them, learning was more efficient and effective, it enhanced learning motivation, and it provided comprehensive accessible learning materials.

Audio-Driven Talking Video Frame Restoration

Talking video frames occasionally drop while streaming for reasons like network errors, which greatly hurts the online team collaboration and user experiences. Directly generating the dropped frames from the remaining ones is unfavorable, since a person's lip motion is usually non-linear and thus hard to be restored when consecutive frames are missing. Nevertheless, the audio content provides strong signals for lip motion and is less likely to drop during transmitting. Inspired by this, as an initial attempt, we present the task of audio-driven talking video frame restoration in this paper, i.e., restoring dropped video frames by jointly leveraging the audio and remaining video frames. Towards the high-quality frame generation, we devise a cross-modal frame restoration network. This network aligns the complete audio content with video frames, precisely identifies and sequentially generates the dropped frames. To justify our model, we construct a new dataset, Talking Video Frames Drop, TVFD for short, consisting of 2.5K video and 144K frames in total. We conduct extensive experiments over TVFD and another publicly accessible dataset - Voxceleb2. Our model obtains significantly improved performance as compared to other state-of-the-art competitors.

Enhancing Location Accuracy by Minimizing RMS Using RSS-AMLE in WSN

Over the past decade, there has been significant growth in wireless sensor networks, particularly in the context of industrial applications. Mobile sensor networks have garnered research interest due to their ability to facilitate communication between various devices. Still, the mobility of these nodes gives rise to challenges such as network coverage and connectivity issues. Addressing these challenges necessitates accurate estimation of sensor node locations, a critical factor in network performance. Numerous methods, such as Angle of Arrival (AOA) and Time of Arrival (TOA), have been proposed for node localization. Still, these methods are plagued by localization errors and high implementation costs. To overcome these localization errors in wireless sensor networks, we present an adaptive approach based on the Received Signal Strength (RSS) model. This model views localization as a non-convex problem and employs an adaptive maximum likelihood estimation to minimize localization errors. An extensive simulation study is carried out to measure the performance of the intended approach in minimizing the localization error. The results unequivocally demonstrate that our localization scheme achieves higher accuracy in locating sensor nodes while reducing deployment costs. Comparative analysis against existing methods further underscores the significance of our approach.

Going Deeper, Generalizing Better: An Information-Theoretic View for Deep Learning.

Deep learning has transformed computer vision, natural language processing, and speech recognition. However, two critical questions remain obscure: 1) why do deep neural networks (DNNs) generalize better than shallow networks and 2) does it always hold that a deeper network leads to better performance? In this article, we first show that the expected generalization error of neural networks (NNs) can be upper bounded by the mutual information between the learned features in the last hidden layer and the parameters of the output layer. This bound further implies that as the number of layers increases in the network, the expected generalization error will decrease under mild conditions. Layers with strict information loss, such as the convolutional or pooling layers, reduce the generalization error for the whole network; this answers the first question. However, algorithms with zero expected generalization error do not imply a small test error. This is because the expected training error is large when the information for fitting the data is lost as the number of layers increases. This suggests that the claim "the deeper the better" is conditioned on a small training error. Finally, we show that deep learning satisfies a weak notion of stability and provides some generalization error bounds for noisy stochastic gradient decent (SGD) and binary classification in DNNs.

Joint Beamforming Design in Reconfigurable Intelligent Surface-Assisted Rate Splitting Networks

Reconfigurable Intelligent Surface (RIS) is an emerging technology that can improve the spectrum and energy efficiency of next-generation wireless networks. However, attaining accurate channel state information (CSI) for the cascaded RIS channel is particularly challenging. Imperfect CSI is a major bottleneck to achieving the spectral efficiency benefit of RIS-assisted networks. Rate splitting (RS), a promising multiple access technology, has been shown to be able to achieve an improved spectrum efficiency and be robust to channel uncertainties. This paper investigates the interplay between RIS and RS by considering a RIS-assisted RS beamforming problem. Active beamforming at the base station (BS) and passive beamforming at the RIS are jointly considered to maximize the minimum user rate under both the perfect CSI case and the imperfect CSI case. A block coordinate descent (BCD) algorithm is developed to solve this non-convex problem. Compared with the conventional semi-definite relaxation (SDR) approach, the proposed method does not require that the covariance matrix of the common beamforming vector be rank-one. Moreover, we present theoretical results to help reveal the impact of the system parameters and explain the performance gain resulting from the integration of RIS and RS. Extensive simulation results are also provided to show that RIS-assisted RS can improve the max-min rate performance significantly compared with conventional multiple access technologies, such as non-orthogonal multiple access (NOMA) and space division multiple access (SDMA) with/without RIS, especially in overloaded systems. With the proposed method, RS shows great potential in combating realistic CSI errors in RIS-assisted networks.

Ignoring Measurement Errors in Social Networks

Abstract We consider peer effect estimation in social network models where some network links are incorrectly measured. We show that if the number or magnitude of mismeasured links does not grow too quickly with the sample size, then standard instrumental variables estimators that ignore these measurement errors remain consistent, and standard asymptotic inference methods remain valid. These results hold even when the link measurement errors are correlated with regressors or with structural errors in the model. Simulations and real data experiments confirm our results in finite samples. These findings imply that researchers can ignore small numbers of mismeasured links in networks.

Academic Experience as Evaluation Material in Online Language Learning

This study aims to explore the phenomena experienced by Lecturers and Students of the Indonesian Language Education Study Program in online learning during the Covid-19 period. The research method is qualitative with a phenomenological approach. Data collection was done through semi-structured interviews with five lecturers and eight students. Interview data was then analyzed within-case and cross-case. Researchers identified two problems related to what lecturers and students experienced in online learning: (1) barriers to using online learning media and (2) obstacles in presenting material and using learning methods. The results showed the lecturers' attitude in carrying out online learning was good by creating a fun online learning atmosphere. Unfortunately, the material provided was not understood by students. They complained about the material and its delivery by the lecturers. It was because the lecturers only provided material and assignments through e-learning. In addition, they only used discussion forums on e-learning to communicate. The researchers also identified that quota and internet network errors were the main problems encountered in online learning. This research can be used as material for evaluating the implementation of online academic activities, especially in language learning of the Indonesian Language Education Study Program. Data validity used member checks, triangulation, and peer debriefing.

Implementation of the SIPASTI BMD Program in Procurement of Goods and Services at the Sumedang Regency Council Secretariat

This article was written to examine the implementation of the SIPASTI BMD program in the procurement of goods and services at the Sumedang Regency Council Secretariat. The method used in preparing this research report is a descriptive qualitative research method. Data collection techniques use literature studies and field studies. In analyzing data, the following steps are used: data reduction, data presentation, conclusion drawing/verification, and triangulation. Based on the research results, it was concluded that the SIPASTI BMD Program was implemented quite well. However, currently, there are several obstacles in its implementation, including network errors that can delay the input of goods and services and some employees who still don't understand the technology. In overcoming these obstacles, several efforts were made, including waiting for policies from the center and improving the quality of employees with better performance, discipline, and following technical guidance or outreach regarding the program. Suggestions related to this research are to fully encourage the Council Secretariat in determining the policies that will be implemented and pay more attention to employees who do not understand technology who will later be used as support in their work as well as facilitating networks at the Council Secretariat to support the SIPASTI BMD program.

Improved thermal hydraulic network modelling and error analysis in disc‐type transformer windings

AbstractTemperature distribution in transformer windings is a crucial factor that influences the safe operation, and lifespan of an insulating system in an oil‐immersed power transformer. Thermal hydraulic network modelling (THNM) has been the subject of numerous studies because of its lower computation cost as compared to a computational fluid dynamics (CFD) simulation. This study proposes improvements to THNM, with primary focus on non‐uniformity of temperatures in the entry and exit vertical oil ducts, and the feasibility of using a piecewise fitting formula for Nusselt number (Nu) of convection heat transfer coefficient in transformer windings. The deviation of hot spot temperature was 0.3°C, and the location of hot spot for CFD and THNM was 34th disc and 35th disc, respectively. The calculation time of CFD and THNM was about 34 h and 35 s under the same computer hardware situations. This provides an acceptable THNM method for the rapid prediction of transformer hot spot temperature and location at low costs for digital twin.

An effective networks intrusion detection approach based on hybrid Harris Hawks and multi-layer perceptron

This paper proposes an Intrusion Detection System (IDS) employing the Harris Hawks Optimization algorithm (HHO) to optimize Multilayer Perceptron learning by optimizing bias and weight parameters. HHO-MLP aims to select optimal parameters in its learning process to minimize intrusion detection errors in networks. HHO-MLP has been implemented using EvoloPy NN framework, an open-source Python tool specialized for training MLPs using evolutionary algorithms. For purposes of comparing the HHO model against other evolutionary methodologies currently available, specificity and sensitivity measures, accuracy measures, and mse and rmse measures have been calculated using KDD datasets. Experiments have demonstrated the HHO MLP method is effective at identifying malicious patterns. HHO-MLP has been tested against evolutionary algorithms like Butterfly Optimization Algorithm (BOA), Grasshopper Optimization Algorithms (GOA), and Black Widow Optimizations (BOW), with validation by Random Forest (RF), XGBoost. HHO-MLP showed superior performance by attaining top scores with accuracy rate of 93.17%, sensitivity level of 89.25%, and specificity percentage of 95.41%.

Short-term Electric Energy Load Forecasting of Ankara Region Using Artificial İntelligence Methods

While energy, which is an indispensable part of everyday life, maintains its importance and place in socio-economic structures of countries, importance of electricity energy, as an important component of energy, is increasing its gravity exponentially. Importance of electricity energy demand maintains its increasing trend in line with growing population, urbanization, industrialization, becoming widespread of technology and welfare. While meeting electricity energy demand, correct and effective load forecasting is needed in order to ensure operational safety without losing the stability of voltage, frequency and power flows within the limits determined under the real time conditions of electric power systems, operate and plan secure and low-cost electricity transmission systems and supply sustainable, reliable, high quality and affordable electricity to consumers. In recent years, use of the artificial intelligence models are quite common in many areas of the power system analysis and control, static and dynamic security analysis, dynamic load modelling and alarm processing and error diagnostics. Therefore, in this study, hybrid artificial neural network model optimized with genetic algorithm has been proposed to perform short-term load forecast since fairly good forecasting results are being obtained especially for non-linear complex problems. The proposed short-term load forecasting model has been tested by making 24 hours load forecasting using actual load data of Ankara Region in Turkey. The results obtained from the proposed model have been compared with classical ANN. In this study, for short-term electric load forecasting based on artificial neural networks and genetic algorithm an adaptive hybrid system has been proposed. Using artificial neural network based on Genetic Algorithm (GA) a new approach for short-term electric load forecasting has been developed. This proposed hybrid algorithm has been implemented for short-term load forecasting. In the proposed model, the actual forecast was made by ANN, Genetic Algorithm is used to select the most appropriate activation function for each node of ANN. Thus, it has been observed that network error decreased even more than classical ANN. The results show that hybrid ANN gives better result than classical ANN. So, the accuracy of the proposed model for short-term load forecasting has been increased. These results showed that the proposed model has a higher accuracy rate than the classical ANN in the short-term load forecasting.

QANS: Toward Quantized Neural Network Adversarial Noise Suppression

Neural network quantization techniques play an important role in efficiently deploying deep learning models on the hardware with limited computing and storage resources. Numerous applications of this technology, such as autopilot, necessitate not just efficiency, but also robustness. Research on the robustness of quantized networks against adversarial attacks is becoming one of the major points of interest. In this work, we rethink the impact of quantization on adversarial attacks and explore the boundary of the robustness of quantized neural networks. This study reveals that activation quantization can be used as a defense to weaken adversarial noise, but the robustness of quantized models is still limited by the amplification effect of network errors, including quantization errors and adversarial noise. To address this problem, we propose the Quantization Adversarial Noise Suppression (QANS) method that employs a Gaussian kernel regularization constraint to stabilize the model by restricting the perturbation error within two levels of tolerance. Extensive experiments are conducted with Wide ResNet and VGG-16 models on CIFAR-10 and SVHN datasets under different attack methods, including several white-box and black-box attacks. Experimental results show the proposed method achieves superior robustness to prior works.

Demand forecasting of cold-chain logistics of aquatic products in China under the background of the Covid-19 post-epidemic era.

In the background of the post-epidemic era, the consumption demand and market scale of cold chain logistics in China are expanding, but there is still an obvious gap with developed countries. To complete the balance between the supply and demand for aquatic products and the rational allocation of logistics resources and promote the rapid development trend of aquatic product cold chain logistics, it is particularly important to forecast and analyze the demand for aquatic product cold chain logistics. This article selects six main factors that affect the demand for aquatic products in cold chain logistics, uses the traditional grey model and the grey-BP neural network model to simulate and predict the demand for aquatic products in cold chain logistics in China from 2012 to 2021, and compares and analyzes the simulation results. Generally speaking, the demand for aquatic products from Chinese residents is on the rise. In the simulation prediction process, the prediction error of the grey-BP neural network is reduced compared to the traditional grey model, and the processing ability of the nonlinear system is ideal. The results show that the grey-BP neural network model is an effective method to predict the demand for cold chain logistics of aquatic products. Finally, suggestions are made on the future development of aquatic cold chain logistics in the post-epidemic era from the economic, social, and environmental aspects, which provide valuable decision-making reference for the development of marine aquaculture enterprises and cold chain logistics industry.

Speech emotion analysis using convolutional neural network (CNN) and gamma classifier-based error correcting output codes (ECOC)

Speech emotion analysis is one of the most basic requirements for the evolution of Artificial Intelligence (AI) in the field of human–machine interaction. Accurate emotion recognition in speech can be effective in applications such as online support, lie detection systems and customer feedback analysis. However, the existing techniques for this field have not yet met sufficient development. This paper presents a new method to improve the performance of emotion analysis in speech. The proposed method includes the following steps: pre-processing, feature description, feature extraction, and classification. The initial description of speech features in the proposed method is done by using the combination of spectro-temporal modulation (STM) and entropy features. Also, a Convolutional Neural Network (CNN) is utilized to reduce the dimensions of these features and extract the features of each signal. Finally, the combination of gamma classifier (GC) and Error-Correcting Output Codes (ECOC) is applied to classify features and extract emotions in speech. The performance of the proposed method has been evaluated using two datasets, Berlin and ShEMO. The results show that the proposed method can recognize speech emotions in the Berlin and ShEMO datasets with an average accuracy of 93.33 and 85.73%, respectively, which is at least 6.67% better than compared methods.

ANALISIS PELAKSANAAN SISTEM INFORMASI REKAM MEDIS BAGIAN PENDAFTARAN RAWAT JALAN DI RSUD PANEMBAHAN SENOPATI BANTUL TAHUN 2019

Background: Medical Record Information System is a very vital requirement for a medical institution. In order to speed up the provision of medical record information services to almost the entire hospital using the medical record information system including the Panembahan General Hospital, Senopati Bantul. Based on the results of a preliminary study conducted on the twenty-first of September 2018 on one of the patients who were waiting in the outpatient registration section with an average waiting time of 11.22 minutes, this complaint was confirmed by 2 other patients who were lining up with these patients.&#x0D; Purpose: Analyze the Implementation of Medical Record Information Systems in the Outpatient Registration Section at Panembahan Senopati Hospital Bantul.&#x0D; Method: This study uses qualitative research methods. This research was carried out by means of source triagulation techniques. This research was carried out in the Outpatient Registration section of Panembahan Senopati Hospital Bantul.&#x0D; Results: The results of the study show that the Analysis of the Medical Record Information System Implementation in Outpatient Registration Section at Panembahan Senopati Hospital Bantul is human resources, facilities and infrastructure, software flow and hardware conditions.&#x0D; Conclusion: Human resources, facilities and infrastructure, software flow and hardware conditions affected by network errors and applications that are down.

Deploying Nagios for Service Monitoring and Reporting

NAGIOS is Server Monitoring tool as an important criteria coming to a data center operation and service .Detects all type of network errors or server issues and allowed to monitor the server performance issues by using Nagios. This can be manually and physically monitoring the location of the server and mapping the network interface and network bridge server will be monitor. Count every 10 minutes once the application will be running properly or else store application related bugs in data center. Large organizations always require fast and efficient network monitoring system which reports to the network administrator as soon as a network problem arises. This paper presents an effective and automatic network monitoring system that continuously monitor all the network switches and inform the administrator by E-MAIL or SMS when any of the network switch goes down. An efficient and automatic network monitoring is always required for large organizations like universities, companies and other business sectors where the manual network monitoring is very difficult. Since large organizations have a big network topology, the manual network monitoring causes waste of time to point out problem location.

Joint optimization strategy of task offloading to mobile edge computing

Offloading strategies in mobile edge computing are hot research, whereas, existing offloading strategies at the edge hard handle the issues of multi-user intensive task scheduling, resulting in the poor utilization of network resource. Therefore, this makes the quality of experience for end users far from satisfactory. To address this, this paper proposes a novel joint offloading strategy consisting of the back propagation neural network and the genetic algorithm. Firstly, using the genetic algorithm optimizes the learning error of the back propagation neural network, and then energy consumption in the system and response delay are jointly optimized by the back propagation neural network. Under long-term total overhead-cost constraints, the joint strategy can achieve the search of the optimal solutions to generate superior calculated offloading results. Unlike those approaches devoting into reducing response delay only for end users, this work takes account into the total overhead-cost in the system thereby affording more efficient for application service providers. Multiple simulation results indicate that the proposed strategy can not only reduce the average response delay of the mobile edge computing system, but also remain a low average energy consumption.