ERROR RESPONSE AND ROBUSTNESS OF A CLASS OF MULTILAYERED PERCEPTRONS WITH THRESHOLD FUNCTIONS

In this paper, based on a stochastic mode! for inputs and weights, and in view of the disturbance of correlative and large input and weight errors, a general algorithm to obtain the output error characteristics of a class of multilayered perceptrons with threshold functions is proposed by using statistical approach. Furthermore, the formula to calculate the robustness of the networks is also given. The result of computer simulation indicates the correctness of the algorithm.

Building up Multi－Layered Perceptrons as Classifier System for Decision Support

This paper focuses on some application issues in m.multi-layered perceptrons researches. The following problem areas are discussed: (1) the classification capability of multi-layered perceptrons; (2) theself-configuration algorithm for facilitating the design of the neural nets' structure;and,finally (3) the application of the fast BP algorithm to speed up the learning procedure. Some experimental results with respect to the application of multi-layered perceptrons as classifier systems in the comprehensive evaluation of Chinese large cities are presented.

MODIFIED OPTIMIZATION LAYER BY LAYER ALGORITHM FOR LEARNING MULTILAYER PERCEPTRONS

Learning of the feedforward multilayer perceptron (MLP) networks is to adapt all synaptic weights in such a way that the discrepancy between the actual output signals and the desired signals, averaged over all learning examples (training patterns), is as small as possible. The backpropagation, or variations thereof, is a standard method applied to adjust the synaptic weights in the network in order to minimize a given cost function. However as a steepest descent approach, BP algorithm is too slow for many applications. Since late 1980s lots of efforts have been reported in the literature aimed at improving the efficiency of the algorithm. Among them a recently proposed learning strategy based on linearization of the nonlinear activation functions and optimization of the multilayer perceptron layer by layer (OLL) seems promising. In this paper a modified learning procedure is presented which tries to find a weight change vector at each trial iteration in the OLL algorithm more efficiently. The proposed learning procedure can save expensive computation efforts and yield better convergence rate as compared to the original OLL learning algorithms especially for large scale networks. The improved OLL learning algorithm is applied to the time series prediction problems presented by the OLL authors, and demonstrates a faster learning capability.

Uncertainties in landslide susceptibility prediction:Influence rule of different levels of errors in landslide spatial position

The accuracy of landslide susceptibility prediction(LSP)mainly depends on the precision of the landslide spatial position.However,the spatial position error of landslide survey is inevitable,resulting in considerable uncertainties in LSP modeling.To overcome this drawback,this study explores the influence of positional errors of landslide spatial position on LSP uncertainties,and then innovatively proposes a semi-supervised machine learning model to reduce the landslide spatial position error.This paper collected 16 environmental factors and 337 landslides with accurate spatial positions taking Shangyou County of China as an example.The 30e110 m error-based multilayer perceptron(MLP)and random forest(RF)models for LSP are established by randomly offsetting the original landslide by 30,50,70,90 and 110 m.The LSP uncertainties are analyzed by the LSP accuracy and distribution characteristics.Finally,a semi-supervised model is proposed to relieve the LSP uncertainties.Results show that:(1)The LSP accuracies of error-based RF/MLP models decrease with the increase of landslide position errors,and are lower than those of original data-based models;(2)70 m error-based models can still reflect the overall distribution characteristics of landslide susceptibility indices,thus original landslides with certain position errors are acceptable for LSP;(3)Semi-supervised machine learning model can efficiently reduce the landslide position errors and thus improve the LSP accuracies.

Multi-layer perceptron-based data-driven multiscale modelling of granular materials with a novel Frobenius norm-based internal variable

One objective of developing machine learning(ML)-based material models is to integrate them with well-established numerical methods to solve boundary value problems(BVPs).In the family of ML models,recurrent neural networks(RNNs)have been extensively applied to capture history-dependent constitutive responses of granular materials,but these multiple-step-based neural networks are neither sufficiently efficient nor aligned with the standard finite element method(FEM).Single-step-based neural networks like the multi-layer perceptron(MLP)are an alternative to bypass the above issues but have to introduce some internal variables to encode complex loading histories.In this work,one novel Frobenius norm-based internal variable,together with the Fourier layer and residual architectureenhanced MLP model,is crafted to replicate the history-dependent constitutive features of representative volume element(RVE)for granular materials.The obtained ML models are then seamlessly embedded into the FEM to solve the BVP of a biaxial compression case and a rigid strip footing case.The obtained solutions are comparable to results from the FEM-DEM multiscale modelling but achieve significantly improved efficiency.The results demonstrate the applicability of the proposed internal variable in enabling MLP to capture highly nonlinear constitutive responses of granular materials.

A hybrid machine learning optimization algorithm for multivariable pore pressure prediction

Pore pressure is essential data in drilling design,and its accurate prediction is necessary to ensure drilling safety and improve drilling efficiency.Traditional methods for predicting pore pressure are limited when forming particular structures and lithology.In this paper,a machine learning algorithm and effective stress theorem are used to establish the transformation model between rock physical parameters and pore pressure.This study collects data from three wells.Well 1 had 881 data sets for model training,and Wells 2 and 3 had 538 and 464 data sets for model testing.In this paper,support vector machine(SVM),random forest(RF),extreme gradient boosting(XGB),and multilayer perceptron(MLP)are selected as the machine learning algorithms for pore pressure modeling.In addition,this paper uses the grey wolf optimization(GWO)algorithm,particle swarm optimization(PSO)algorithm,sparrow search algorithm(SSA),and bat algorithm(BA)to establish a hybrid machine learning optimization algorithm,and proposes an improved grey wolf optimization(IGWO)algorithm.The IGWO-MLP model obtained the minimum root mean square error(RMSE)by using the 5-fold cross-validation method for the training data.For the pore pressure data in Well 2 and Well 3,the coefficients of determination(R^(2))of SVM,RF,XGB,and MLP are 0.9930 and 0.9446,0.9943 and 0.9472,0.9945 and 0.9488,0.9949 and 0.9574.MLP achieves optimal performance on both training and test data,and the MLP model shows a high degree of generalization.It indicates that the IGWO-MLP is an excellent predictor of pore pressure and can be used to predict pore pressure.

Dynamic Multi-Layer Perceptron for Fetal Health Classification Using Cardiotocography Data

Fetal health care is vital in ensuring the health of pregnant women and the fetus.Regular check-ups need to be taken by the mother to determine the status of the fetus’growth and identify any potential problems.To know the status of the fetus,doctors monitor blood reports,Ultrasounds,cardiotocography(CTG)data,etc.Still,in this research,we have considered CTG data,which provides information on heart rate and uterine contractions during pregnancy.Several researchers have proposed various methods for classifying the status of fetus growth.Manual processing of CTG data is time-consuming and unreliable.So,automated tools should be used to classify fetal health.This study proposes a novel neural network-based architecture,the Dynamic Multi-Layer Perceptron model,evaluated from a single layer to several layers to classify fetal health.Various strategies were applied,including pre-processing data using techniques like Balancing,Scaling,Normalization hyperparameter tuning,batch normalization,early stopping,etc.,to enhance the model’s performance.A comparative analysis of the proposed method is done against the traditional machine learning models to showcase its accuracy(97%).An ablation study without any pre-processing techniques is also illustrated.This study easily provides valuable interpretations for healthcare professionals in the decision-making process.

Enhancing Healthcare Data Security and Disease Detection Using Crossover-Based Multilayer Perceptron in Smart Healthcare Systems

The healthcare data requires accurate disease detection analysis,real-timemonitoring,and advancements to ensure proper treatment for patients.Consequently,Machine Learning methods are widely utilized in Smart Healthcare Systems(SHS)to extract valuable features fromheterogeneous and high-dimensional healthcare data for predicting various diseases and monitoring patient activities.These methods are employed across different domains that are susceptible to adversarial attacks,necessitating careful consideration.Hence,this paper proposes a crossover-based Multilayer Perceptron(CMLP)model.The collected samples are pre-processed and fed into the crossover-based multilayer perceptron neural network to detect adversarial attacks on themedical records of patients.Once an attack is detected,healthcare professionals are promptly alerted to prevent data leakage.The paper utilizes two datasets,namely the synthetic dataset and the University of Queensland Vital Signs(UQVS)dataset,from which numerous samples are collected.Experimental results are conducted to evaluate the performance of the proposed CMLP model,utilizing various performancemeasures such as Recall,Precision,Accuracy,and F1-score to predict patient activities.Comparing the proposed method with existing approaches,it achieves the highest accuracy,precision,recall,and F1-score.Specifically,the proposedmethod achieves a precision of 93%,an accuracy of 97%,an F1-score of 92%,and a recall of 92%.

Recommendation System Based on Perceptron and Graph Convolution Network

The relationship between users and items,which cannot be recovered by traditional techniques,can be extracted by the recommendation algorithm based on the graph convolution network.The current simple linear combination of these algorithms may not be sufficient to extract the complex structure of user interaction data.This paper presents a new approach to address such issues,utilizing the graph convolution network to extract association relations.The proposed approach mainly includes three modules:Embedding layer,forward propagation layer,and score prediction layer.The embedding layer models users and items according to their interaction information and generates initial feature vectors as input for the forward propagation layer.The forward propagation layer designs two parallel graph convolution networks with self-connections,which extract higher-order association relevance from users and items separately by multi-layer graph convolution.Furthermore,the forward propagation layer integrates the attention factor to assign different weights among the hop neighbors of the graph convolution network fusion,capturing more comprehensive association relevance between users and items as input for the score prediction layer.The score prediction layer introduces MLP(multi-layer perceptron)to conduct non-linear feature interaction between users and items,respectively.Finally,the prediction score of users to items is obtained.The recall rate and normalized discounted cumulative gain were used as evaluation indexes.The proposed approach effectively integrates higher-order information in user entries,and experimental analysis demonstrates its superiority over the existing algorithms.

Machine learning methods for predicting CO_(2) solubility in hydrocarbons

The application of carbon dioxide(CO_(2)) in enhanced oil recovery(EOR) has increased significantly, in which CO_(2) solubility in oil is a key parameter in predicting CO_(2) flooding performance. Hydrocarbons are the major constituents of oil, thus the focus of this work lies in investigating the solubility of CO_(2) in hydrocarbons. However, current experimental measurements are time-consuming, and equations of state can be computationally complex. To address these challenges, we developed an artificial intelligence-based model to predict the solubility of CO_(2) in hydrocarbons under varying conditions of temperature, pressure, molecular weight, and density. Using experimental data from previous studies,we trained and predicted the solubility using four machine learning models: support vector regression(SVR), extreme gradient boosting(XGBoost), random forest(RF), and multilayer perceptron(MLP).Among four models, the XGBoost model has the best predictive performance, with an R^(2) of 0.9838.Additionally, sensitivity analysis and evaluation of the relative impacts of each input parameter indicate that the prediction of CO_(2) solubility in hydrocarbons is most sensitive to pressure. Furthermore, our trained model was compared with existing models, demonstrating higher accuracy and applicability of our model. The developed machine learning-based model provides a more efficient and accurate approach for predicting CO_(2) solubility in hydrocarbons, which may contribute to the advancement of CO_(2)-related applications in the petroleum industry.

Bimodal Coupling Haptic Perceptron for Accurate Contactless Gesture Perception and Material Identification

A bimodal coupled multifunctional tactile perceptron for contactless gesture recognition and material identification is pro-posed to address the challenges posed by limited functionality,signal interference from multimodal collaborative work,and the high power consumption of traditional tactile sensors.This perceptron integrates a capacitive sensor and a triboelectric sensor symmetrically,employing an energy complementarity strategy to reduce power consumption and implementing symmetrical distribution of two sensors for physical isolation to prevent signal interference.The capacitive sensor detects external pressure,providing information on material properties such as hardness,softness,and deformation,with a wide linear response range of 0-745.3 kPa.The triboelectric sensor captures the electron affinity of measured object.Further,by utilising machine learning algorithms,a system for contactless gesture recognition and material identification is engineered.This system demonstrates a remarkable accuracy rate of 98.5%when recognising 5 gestures,and achieves a perfect identi-fication(100%)of 10 different materials aided by incorporating capacitive and triboelectric response.These results greatly advance the progress of tactile perceptrons with high integration,low power consumption,and multifunctionality,enhancing their effectiveness and reliability in smart device applications.

Improved reservoir characterization by means of supervised machine learning and model-based seismic impedance inversion in the Penobscot field,Scotian Basin

The present research work attempted to delineate and characterize the reservoir facies from the Dawson Canyon Formation in the Penobscot field,Scotian Basin.An integrated study of instantaneous frequency,P-impedance,volume of clay and neutron-porosity attributes,and structural framework was done to unravel the Late Cretaceous depositional system and reservoir facies distribution patterns within the study area.Fault strikes were found in the EW and NEE-SWW directions indicating the dominant course of tectonic activities during the Late Cretaceous period in the region.P-impedance was estimated using model-based seismic inversion.Petrophysical properties such as the neutron porosity(NPHI)and volume of clay(VCL)were estimated using the multilayer perceptron neural network with high accuracy.Comparatively,a combination of low instantaneous frequency(15-30 Hz),moderate to high impedance(7000-9500 gm/cc*m/s),low neutron porosity(27%-40%)and low volume of clay(40%-60%),suggests fair-to-good sandstone development in the Dawson Canyon Formation.After calibration with the welllog data,it is found that further lowering in these attribute responses signifies the clean sandstone facies possibly containing hydrocarbons.The present study suggests that the shale lithofacies dominates the Late Cretaceous deposition(Dawson Canyon Formation)in the Penobscot field,Scotian Basin.Major faults and overlying shale facies provide structural and stratigraphic seals and act as a suitable hydrocarbon entrapment mechanism in the Dawson Canyon Formation's reservoirs.The present research advocates the integrated analysis of multi-attributes estimated using different methods to minimize the risk involved in hydrocarbon exploration.

Pioneering role of machine learning in unveiling intensive care unitacquired weakness

In the research published in the World Journal of Clinical Cases,Wang and Long conducted a quantitative analysis to delineate the risk factors for intensive care unit-acquired weakness(ICU-AW)utilizing advanced machine learning methodologies.The study employed a multilayer perceptron neural network to accurately predict the incidence of ICU-AW,focusing on critical variables such as ICU stay duration and mechanical ventilation.This research marks a significant advancement in applying machine learning to clinical diagnostics,offering a new paradigm for predictive medicine in critical care.It underscores the importance of integrating artificial intelligence technologies in clinical practice to enhance patient management strategies and calls for interdisciplinary collaboration to drive innovation in healthcare.

Development of a Deep Learning Model for the Prognosis of the Occurrence of Death from Stomach Cancer in Senegal

Context and Objectives: Stomach cancer ranks fifth in incidence and fourth in mortality worldwide. In Senegal, there were 597 new cases in 2020, with a mortality rate of almost 70%. The aim of this study was to develop a machine-learning model for the prognosis of death from stomach cancer 5 years after treatment. Methods: Our study sample consisted of 262 patients treated for gastric cancer at Aristide le Dantec Hospital and followed postoperatively between 2007 and 2020. We developed a multilayer perceptron with optimal hyperparameters and compared its performance with standard classification algorithms. We also augmented our data with a set of synthetic data generators to evaluate the behaviour of the model when faced with a larger amount of data. Results: Our model obtained an accuracy of 97.5%, outperforming the SVM (93%), RF (93.8%) and KNN (92.7%) models. An improvement of 1.5% in accuracy was achieved with synthetic data. Our study showed that the most pejorative factors in the evolution of the cancer were the appearance of hepatic metastases or adenopathy, smoking, and the infiltrative and stenosing aspects of the tumour on endoscopy. Conclusion: Our model predicted the occurrence of death from gastric cancer with very high accuracy, outperforming standard classification algorithms. The increase in training data produced an improvement in accuracy. Our study will help doctors to personalize the management of gastric cancer patients.

People Recognition by RGB and NIR Analysis from Digital Image Database Using Cross-Correlation and Wavelets

This document presents a framework for recognizing people by palm vein distribution analysis using cross-correlation based signatures to obtain descriptors. Haar wavelets are useful in reducing the number of features while maintaining high recognition rates. This experiment achieved 97.5% of individuals classified correctly with two levels of Haar wavelets. This study used twelve-version of RGB and NIR (near infrared) wavelength images per individual. One hundred people were studied;therefore 4,800 instances compose the complete database. A Multilayer Perceptron (MLP) was trained to improve the recognition rate in a k-fold cross-validation test with k = 10. Classification results using MLP neural network were obtained using Weka (open source machine learning software).

A Multilayer Perceptron Artificial Neural Network Study of Fatal Road Traffic Crashes

This paper examines the relationship between fatal road traffic accidents and potential predictors using multilayer perceptron artificial neural network (MLANN) models. The initial analysis employed twelve potential predictors, including traffic volume, prevailing weather conditions, roadway characteristics and features, drivers’ age and gender, and number of lanes. Based on the output of the model and the variables’ importance factors, seven significant variables are identified and used for further analysis to improve the performance of models. The model is optimized by systematically changing the parameters, including the number of hidden layers and the activation function of both the hidden and output layers. The performances of the MLANN models are evaluated using the percentage of the achieved accuracy, R-squared, and Sum of Square Error (SSE) functions.

基于Perceptron建立某车型焊钉及白车身在线检测体系

以北京奔驰汽车有限公司206车型为基础,基于Perceptron就螺柱焊钉位置尺寸、车身尺寸和覆盖件尺寸在线自动检测以及相应的精度控制进行了研究,规划并建立焊钉位置尺寸在线测量点76个,普通测量特征1373个测量点;规划并实施焊钉在线测量方案以及后端、Z1和Z2.3在线测量方案;完成在线检测工位测量工装优化改进,达成了基于Perceptron的V206焊钉及白车身在线检测体系建设,并实现了工时和成本的节约。

基于多层SimHash的Android恶意应用程序检测方法

提出一个基于多层SimHash的相似度检测方法,通过对APK文件进行分析,最终从5个方面提取分析内容来表征APK,同时在每一层上使用改进的SimHash方法进行相似度检测分析。通过从APK文件中提取的Android Manifest.xml文件、从dex反编译得出的Smali代码累加和、Smali文件指令提取、Java代码集合、Java指令集提取5个层面进行分析。同时通过学习Voted Perceptron投票算法,将其应用到检测过程中,采用信任值权重的方法,为每一层赋予一个可信值,并在最后得出结果时将每一层结果结合权重分析,实验分析结果表明该方法具有更好的检测效果。

Wireless location algorithm using digital broadcasting signals based on neural network

In order to enhance the accuracy and reliability of wireless location under non-line-of-sight （NLOS） environments,a novel neural network （NN） location approach using the digital broadcasting signals is presented. By the learning ability of the NN and the closely approximate unknown function to any degree of desired accuracy,the input-output mapping relationship between coordinates and the measurement data of time of arrival （TOA） and time difference of arrival （TDOA） is established. A real-time learning algorithm based on the extended Kalman filter （EKF） is used to train the multilayer perceptron （MLP） network by treating the linkweights of a network as the states of the nonlinear dynamic system. Since the EKF-based learning algorithm approximately gives the minimum variance estimate of the linkweights,the convergence is improved in comparison with the backwards error propagation （BP） algorithm. Numerical results illustrate thatthe proposedalgorithmcanachieve enhanced accuracy,and the performance ofthe algorithmis betterthanthat of the BP-based NN algorithm and the least squares （LS） algorithm in the NLOS environments. Moreover,this location method does not depend on a particular distribution of the NLOS error and does not need line-of-sight （ LOS ） or NLOS identification.

Performance evaluation of series and parallel strategies for financial time series forecasting

Background:Improving financial time series forecasting is one of the most challenging and vital issues facing numerous financial analysts and decision makers.Given its direct impact on related decisions,various attempts have been made to achieve more accurate and reliable forecasting results,of which the combining of individual models remains a widely applied approach.In general,individual models are combined under two main strategies:series and parallel.While it has been proven that these strategies can improve overall forecasting accuracy,the literature on time series forecasting remains vague on the choice of an appropriate strategy to generate a more accurate hybrid model.Methods:Therefore,this study’s key aim is to evaluate the performance of series and parallel strategies to determine a more accurate one.Results:Accordingly,the predictive capabilities of five hybrid models are constructed on the basis of series and parallel strategies compared with each other and with their base models to forecast stock price.To do so,autoregressive integrated moving average(ARIMA)and multilayer perceptrons(MLPs)are used to construct two series hybrid models,ARIMA-MLP and MLP-ARIMA,and three parallel hybrid models,simple average,linear regression,and genetic algorithm models.Conclusion:The empirical forecasting results for two benchmark datasets,that is,the closing of the Shenzhen Integrated Index(SZII)and that of Standard and Poor’s 500(S&P 500),indicate that although all hybrid models perform better than at least one of their individual components,the series combination strategy produces more accurate hybrid models for financial time series forecasting.