Multiple Multilayer Perceptron Research Articles

Multi-Output Prediction Model for Basic Oxygen Furnace Steelmaking Based on the Fusion of Deep Convolution and Attention Mechanisms

The objective of basic oxygen furnace (BOF) steelmaking is to achieve molten steel with final carbon content, temperature, and phosphorus content meeting the requirements. Accurate prediction of the above properties is crucial for end-point control in BOF steelmaking. Traditional prediction models typically use multi-variable input and single-variable output approaches, neglecting the coupling relationships between different property indicators, making it difficult to predict multiple outputs simultaneously. Consequently, a multi-output prediction model based on the fusion of deep convolution and attention mechanism networks (FDCAN) is proposed. The model inputs include scalar data, such as the properties of raw materials and target molten steel, and time series data, such as lance height, oxygen supply intensity, and bottom air supply intensity during the blowing process. The FDCAN model utilizes a fully connected module to extract nonlinear features from scalar data and a deep convolution module to process time series data, capturing high-dimensional feature representations. The attention mechanism then assigns greater weight to significant features. Finally, multiple multi-layer perceptron modules predict the outputs—final carbon content, temperature, and phosphorus content. This structure allows FDCAN to learn complex relationships within the input data and between input and output variables. The effectiveness of the FDCAN model is validated using actual BOF steelmaking data, achieving hit rates of 95.14% for final carbon content within ±0.015 wt%, 84.72% for final temperature within ±15 °C, and 88.89% for final phosphorus content within ±0.005 wt%.

Improving multi-site photovoltaic forecasting with relevance amplification: DeepFEDformer-based approach

The present research on photovoltaic (PV) forecasting is devoted to the use of spatial information about PV sites to improve the accuracy of the models, but most of the models have to increase their network complexity to learn the spatial dependence. In this paper, we propose to take advantage of the known geographic location information of PV sites and embed them directly into the input information of Decoder, which makes it easier for the model to focus its attention. We refer to this process as relevance amplification. Based on this, this paper proposes Relevance Amplification based DeepFEDformer (RAD-FEDformer), where DeepFEDformer adds multiple Multi-Layer Perceptron (MLP) layers to FEDformer to improve the perception of deep features. The Relevance Amplification Module (RAM) is designed to receive geographic correlation information as a way to enhance its influence on the Seasonality component of the Decoder input and improve the performance of the attention mechanism in the model. Using the power generation data from PV plants distributed in 11 regions of Belgium as our case study, we evaluated the performance of RAD-FEDformer in predicting PV data at 48/96/192 time steps into the future for each plant. Compared to other Transformer family models, RAD-FEDformer achieved SOTA results by demonstrating significant R2 improvements of 16.70%, 51.45%, and 23.50% over FEDformer, along with an average MSE reduction of 16.9%. We designed Ablation Experiments to validate and compare the performance of MLPs of different sizes based on the ETTm1 dataset and discussed the performance of RAM and its effect on the attention mechanism based on the PV dataset. The results show that our model enhances the performance of the attention mechanism in Multi-Site PV prediction scenarios, and conclude that the optimization of RAM is more effective in longer sequence predictions.

Can Machine Learning Predict Running Kinematics Based on Upper Trunk GPS-Based IMU Acceleration? A Novel Method of Conducting Biomechanical Analysis in the Field Using Artificial Neural Networks

This study aimed to investigate whether running kinematics can be accurately estimated through an artificial neural network (ANN) model containing GPS-based accelerometer variables and anthropometric data. Thirteen male participants with extensive running experience completed treadmill running trials at several speeds. Participants wore a GPS device containing a triaxial accelerometer, and running kinematics were captured by an 18-camera motion capture system for each trial. Multiple multilayer perceptron neural network models were constructed to estimate participants’ 3D running kinematics. The models consisted of the following input variables: 3D peak accelerometer acceleration during foot stance (g), stance time (s), running speed (km/h), participant height (cm), leg length (cm), and mass (kg). Pearson’s correlation coefficient (r), root mean squared error (RMSE), and relative root mean squared error (rRMSE) showed that ANN models provide accurate estimations of joint/segment angles (mean rRMSE = 13.0 ± 4.3%) and peak segment velocities (mean rRMSE = 22.1 ± 14.7%) at key gait phases across foot stance. The highest accuracies were achieved for flexion/extension angles of the thorax, pelvis, and hip, and peak thigh flexion/extension and vertical velocities (rRMSE < 10%). The current findings offer sports science and medical practitioners working with this data a method of conducting field-based analyses of running kinematics using a single IMU.

Machine learning tabulation of thermochemistry for turbulent dimethyl ether (DME) flames

The present work applies a machine learning tabulation methodology for the thermochemistry of dimethyl ether (DME), a typical biodiesel fuel, to accelerate the real-time computation of a large-size DME mechanism in turbulent non-premixed combustion. This approach is known as the hybrid flamelet/random data and multiple multilayer perceptrons (HFRD-MMLP) method (Ding et al., 2021). The essence of the HFRD-MMLP method lies in the generation of training data using the HFRD approach, which enhances the capacity of generalisation for the reactive composition space encountered in practical turbulent combustion by using the random process to expand the training dataset from laminar flamelets. The MMLP artificial neural networks (ANNs) are then trained to predict different composition states, aiming to improve the accuracy of ANN predictions. To validate the effectiveness of the ANNs, they are initially tested on 1-D laminar flame simulations with varying strain rates. Subsequently, a test is conducted on Sandia non-premixed turbulent DME series flames D and F, with an increasing jet Reynolds Number. The results regarding species mole fractions and temperature show overall excellent agreement with the direct integration method. The HFRD-MMLP method achieves speed-up factors of over 16 and 10 for the reaction step, and total computational costs, respectively, in the LES-PDF simulation with the DME mechanism in this work, surpassing the speed-up factors of approximately 12–14, and below 5 for the reaction step, and total time costs, respectively, in previous works on GRI-1.2 mechanism for CH4 (Ding et al., 2021) and CH4/H2 (Ding et al., 2022) combustion. This indicates that the HRFD-MMLP method is highly effective for mechanisms of large size in reducing the computational cost by avoiding a significant number of highly stiff ordinary differential equations (ODEs) integration. This approach can be applied in real-time calculations of reaction source terms in methods including Direct Numerical Simulation (DNS), Probability Density Function (PDF) methods, unsteady flamelet, Conditional Moment Closure (CMC), Multiple Mapping Closure (MMC), Linear Eddy Model (LEM), Thickened Flame Model, the Partially Stirred Reactor (PaSR) method (as in OpenFOAM) and laminar flame computation.

Simulation of turbulent premixed flames with machine learning - tabulated thermochemistry

The numerical integration of the differential equations describing chemical kinetics consumes the majority of computational time in combustion simulations that involve direct coupling of chemistry and flow, such as transported probability density function (PDF) methods, direct numerical simulation (DNS), conditional moment closure (CMC), unsteady flamelet, multiple mapping closure (MMC), thickened flame model, linear eddy model (LEM), partially stirred reactor (PaSR) as in OpenFOAM and laminar flame computation. This step can be accelerated by tabulation, and artificial neural networks (ANNs) have recently emerged as a powerful technique in this domain. To be applicable to a wide family of problems, an ANN tabulation approach must be based on data generated by an abstract process, rather than from the turbulent flame to be simulated. In the present work, the hybrid flamelet/random data and multiple multilayer perceptrons (HFRD-MMLP) method (Ding et al., Combust. Flame 231, 111493, 2021) for non-premixed flames is taken as a basis to develop a thermochemistry tabulation method for premixed flames. In the spirit of maintaining an essentially random data set that still originates in meaningful composition states, a set of one-dimensional premixed flame simulations is employed to generate data that are used as starting points for a random data generation process and subsequently discarded. The approach is applied to large eddy simulations (LES) of the Cambridge/Sandia swirl burner in configurations five and six, with the transported PDF method employed to provide closure for the filtered reaction source terms and the stochastic fields method used for numerical solution. Very good agreement in both major and minor species is observed between the LES-PDF simulations using direct integration of the reaction source term and the ones with the ANNs. Furthermore, the average time taken for reaction source term computations is reduced by fourteen times, while memory requirements constitute only 1.4 MB.

Accurate Prediction and Reliable Parameter Optimization of Neural Network for Semiconductor Process Monitoring and Technology Development

Herein, novel neural network (NN) methods that improve prediction accuracy and reduce output variance of the optimized input in the gradient method for cross‐sectional data are proposed, and the variability evaluation approach of optimized inputs in the semiconductor process is suggested. Specifically, electrical parameter measurements (EPMs) and power‐delay product of industrial high‐k metal gate DRAM peripheral 29‐stage ring oscillator circuits, including NMOS, PMOS, and interconnects, are focused on. The proposed methods find an optimized input to achieve a lower NN output variance in the gradient descent than one multilayer perceptron (MLP) and mean ensemble of MLPs even when considering the variabilities of the devices and interconnects. The local optima problem of one MLP is resolved by utilizing multiple MLPs trained with different train/validation data, their trimmed mean, and an additional learnable layer. Moreover, adding the learnable layer secures versatility for various parametric datasets. The methods improve the prediction accuracy (R2) by 5.6–15.6% in sparse data space compared to one MLP and the mean ensemble, decrease the NN output variance of the optimized input by 73.0–81.6% compared to one MLP and the mean ensemble, and are successfully verified by implementing it on EPMs of 3977 test patterns of 314 wafers and 16 lots.

Multilayer perceptron neural network with regression and ranking loss for patient-specific quality assurance

Patient-specific quality assurance (PSQA) of volumetric modulated arc therapy (VMAT) treatment plans is crucial to enable the plans to be validated for clinical acceptance. However, performing PSQA for clinical delivery is labor-intensive and time-consuming. The existing prediction models do not take into account the dynamic delivery process of VMAT plans. To solve the above problems and improve accuracy of PSQA, this paper presents a multilayer perceptron (MLP) neural network model with regression and ranking loss to predict the gamma passing rate (GPR). The proposed model combines a convolutional neural network with multiple MLP blocks for extracting inter-image correlation features of plan files during dynamic delivery. To focus on the similarity and specificity of multiple VMAT plans, a regression and ranking loss function with dynamic weights is proposed to optimize the training process. In addition, a clinical workflow is proposed to combine the designed model with measurement-based PSQA to screen potential risky plans better. A total of 690 VMAT plans from multiple treatment sites are collected to validate the performance. For 2%/2 mm, 3%/2 mm and 3%/3 mm, the best result of mean absolute error and max error between measured and predicted GPR are 2.17%, 1.25%, 0.74%, and 7.89%, 4.29%, 3.05%, respectively. Experimental results demonstrate that the proposed method has a state-of-the-art performance and can improve the VMAT PSQA process and reduce PSQA workloads.

ICE: Implicit Coordinate Encoder for Multiple Image Neural Representation.

In recent years, implicit neural representations (INR) have shown their great potential to solve many computer graphics and computer vision problems. With this technique, signals such as 2D images or 3D shapes can be fit by training multi-layer perceptrons (MLP) on continuous functions, providing many advantages over conventional discrete representations. Despite being considered a promising approach to 2D image encoding and compression, the application of INR to image collections remains a challenge, since the number of parameters needed rapidly grow with the number of images. In this paper, we propose a fully implicit approach to INR which drastically reduces the size of MLP models in multiple image representation tasks. We introduce the concept of implicit coordinate encoder (ICE) and show it can be used to scale INR with the image number; specifically, by learning a common feature space between images. Furthermore, we show that our method is valid not only for image collections but also for large (gigapixel) images by applying a "divide-and-conquer" strategy. We propose an auto-encoder deep neural network architecture, with a single ICE (encoder) and multiple MLP (decoders), which are jointly trained following a multi-task learning strategy. We demonstrate the benefits coming from ICE when it is implemented as a one-dimensional convolutional encoder, including a better performance of the downstream MLP models with an order of magnitude fewer parameters. Our method is the first one to make use of convolutional blocks in INR networks, unlike the conventional approach of using MLP architectures only. We show the benefits of ICE in two experimental scenarios: a collection of twenty-four small ( 768×512 ) images (Kodak dataset), and a single large ( 3072×3072 ) image (dwarf planet Pluto), achieving better quality than previous fully-implicit methods, using up to 50% fewer parameters.

Offline reinforcement learning control for electricity and heat coordination in a supercritical CHP unit

With a high proportion of renewable energy generation connected to the power grid, the combined heat and power (CHP) units need to have flexible operation and control capabilities over a wide range of variable load conditions. In China, the proportion of supercritical combined heat and power (S-CHP) units gradually increases due to their high thermal efficiency. In this paper, we propose a data-driven environment modeling method and an offline reinforcement learning-based electricity–heat coordinated control approach for the wide and flexible load adjustment capabilities of the S-CHP unit. First of all, a modeling method based on the multiple multilayer perceptron (MLP) ensemble is proposed to address the possible over-fitting produced by a single MLP. Then, the state and reward are set considering the dynamic characteristics of the S-CHP unit. Moreover, policy training is implemented through a soft actor-critic algorithm with a maximum entropy model to ensure more robust search capabilities under variable load conditions and targets. The simulation results show that the generalization ability of the environment in the multiple MLP ensemble mode is more substantial than that in the single MLP mode. In addition, when the electric load command is between 267 MW and 325 MW, the offline reinforcement learning can significantly reduce the integral of absolute error matrices of the output parameters, demonstrating that the proposed strategy can achieve electricity–heat coordinated control under variable load conditions.

MLP-TLBO: Combining Multi-Layer Perceptron Neural Network and Teaching-Learning-Based Optimization for Breast Cancer Detection

Symptoms and types of breast cancer, like most other diseases, vary from person to person. This is while some infected people may not have any symptoms. Breast cancer is one of the common cancers among women, of course, some types of it also affect men. Basically, timely and early diagnosis before the progress of the disease can significantly provide treatment grounds. There are many approaches to breast cancer diagnosis. Classification of patients in terms of benign or malignant cancer using data mining techniques has attracted the attention of many researchers. Data mining can provide a medical assistant system with very low cost and high accuracy for early detection of breast cancer. In this regard, this study proposes a data mining-based method by a combining Multi-Layer Perceptron (MLP) neural network and an evolutionary approach for breast cancer diagnosis. We aim to tune MLP parameters using an evolutionary approach. Here, Teaching-Learning-Based Optimization (TLBO) is used as a new evolutionary approach. However, the capabilities of approaches such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO) and Open-Source Development Model Algorithm (ODMA) are analyzed compared to TLBO. The proposed method named MLP-TLBO can simultaneously adjust all MLP parameters (i.e., effective features and their number, number of hidden layers, number of neurons in each hidden layer, and weights of links). To improve the classification process, MLP-TLBO develops an ensemble classification mechanism in which multiple MLPs simultaneously model the training data. The evaluation of the proposed method is done on the Wisconsin Breast Cancer Database. We use three common datasets from Wisconsin including WBCD, WDBC and WPBC for simulations. Based on the results, the proposed method has promising results and averages between 7% and 26% better performance compared to similar methods.

Machine learning tabulation of thermochemistry of fuel blends

The objective of the present work is to develop a machine learning tabulation methodology for thermochemistry that accounts for fuel blends. The approach is based on the hybrid flamelet/random data and multiple multilayer perceptrons (HFRD-MMLP) methodology (Ding et al., 2021), the essence of which is to train a set of artificial neural networks (ANNs) using random data so as to anticipate the composition space encountered in turbulent flame simulations. As such, it is applicable to any combustion modelling approach that involves direct coupling of chemistry and flow, such as transported probability density function (PDF) methods, direct numerical simulation (DNS), conditional moment closure (CMC), unsteady flamelet, multiple mapping closure (MMC), thickened flame model, linear eddy model (LEM), partially stirred reactor (PaSR) as in OpenFOAM and laminar flame computation. In this paper, the HFRD approach is further developed to generate data of varying fuel ratios. Furthermore, radiative heat losses are included and it is shown that the ANN-based simulations are able to account for it. The ANNs generated are first tested on 1-D laminar flame simulations and then applied to two turbulent flames with different fuel compositions: a pure methane flame, Sandia flame D, and Sydney flame HM1, which is a methane/hydrogen flame. The results of species mass fraction and temperature are compared between ANN and direct integration, and excellent agreement are achieved. These results indicate that the methodology has great capacity for generalisation and is applicable to a range of blended fuels. Furthermore, a speed-up ratio of 14 to 17 is attained for the reaction step compared with direct integration, which greatly reduces the computational cost of turbulent combustion simulations.

Predicting crop yields using a new robust Bayesian averaging model based on multiple hybrid ANFIS and MLP models

Predicting crop yield is an important issue for farmers. Food security is important for decision-makers. The agriculture industry can more accurately supply human demand for food if the crop yield is predicted accurately. Tomato is one of the most important crops so that 160 million tonnes of tomatoes are produced annually around the world. In this study, tomato yield based on data of 40 cities of Iran country including annual average temperature (T), relative humidity (RH), effective rainfall (R), wind speed (WS), and Evapotranspiration (EV) for the period of 1968–2018 was predicted using a new Bayesian model averaging (BMA). The paper's main innovation is the use of the new BMA so that it allows the modellers to quantify the uncertainty of model parameters and inputs simultaneously. For this aim, first, the multiple Adaptive neuro-fuzzy interface system (ANFIS) and multi-layer perceptron (MLP) were used for predicting crop yield. To train the ANFIS and MLP model, a new algorithm, namely, multi verse optimization algorithm (MOA) was used. Also, the ability of MOA was benchmarked against the particle swarm optimization (PSO), and firefly algorithm (FFA). In the next level, the new BMA used the outputs of the ANFIS-MOA, MLP-MOA, ANFIS, FFA, MLP-FFA, ANFIS-PSO, MLP-PSO, ANFIS, and MLP for predicting tomato yield in an ensemble framework. The five- input combination of RH, T, and R, WS, and EV gave the best result. The mean absolute error (MAE) of the BMA in the testing level was 20.12 (Ton/ha) while it was 24.12, 24.45, 24.67, 25.12, 29.12, 30.12, 31.12, and 33.45 for the ANFIS-MOA, MLP-MOA, ANFIS-FFA, MLP-FFA, ANFIS-PSO, MLP-PSO, ANFIS, and MLP models. Regarding the results of uncertainty analysis, the uncertainty of BMA was lower than those of the ANFIS-MOA, MLP-MOA, ANFIS-FFA, MLP-FFA, ANFIS-PSO, MLP-PSO, ANFIS, and MLP models while the MLP model provided the highest uncertainty. The results of this study indicated that BMA using multiple MLP and ANFIS model was useful for predicting tomato yield.

Combined Multi-Atlas and Multi-Layer Perception for Alzheimer's Disease Classification.

Alzheimer's disease (AD) is a progressive and irreversible neurodegenerative disease. To distinguish the stage of the disease, AD classification technology challenge has been proposed in Pattern Recognition and Computer Vision 2021 (PRCV 2021) which provides the gray volume and average cortical thickness data extracted in multiple atlases from magnetic resonance imaging (MRI). Traditional methods either train with convolutional neural network (CNN) by MRI data to adapt the spatial features of images or train with recurrent neural network (RNN) by temporal features to predict the next stage. However, the morphological features from the challenge have been extracted into discrete values. We present a multi-atlases multi-layer perceptron (MAMLP) approach to deal with the relationship between morphological features and the stage of the disease. The model consists of multiple multi-layer perceptron (MLP) modules, and morphological features extracted from different atlases will be classified by different MLP modules. The final vote of all classification results obtains the predicted disease stage. Firstly, to preserve the diversity of brain features, the most representative atlases are chosen from groups of similar atlases, and one atlas is selected in each group. Secondly, each atlas is fed into one MLP to fetch the score of the classification. Thirdly, to obtain more stable results, scores from different atlases are combined to vote the result of the classification. Based on this approach, we rank 10th among 373 teams in the challenge. The results of the experiment indicate as follows: (1) Group selection of atlas reduces the number of features required without reducing the accuracy of the model; (2) The MLP architecture achieves better performance than CNN and RNN networks in morphological features; and (3) Compared with other networks, the combination of multiple MLP networks has faster convergence of about 40% and makes the classification more stable.

Side-channel Attack Using Word Embedding and Long Short Term Memories

Side-channel attack (SCA) based on machine learning has proved to be a valid technique in cybersecurity, especially subjecting to the symmetric-key crypto implementations in serial operation. At the same time, parallel-encryption computing based on Field Programmable Gate Arrays (FPGAs) grows into a new influencer, but the attack results using machine learning are exiguous. Research on the traditional SCA has been mostly restricted to pre-processing: Signal Noisy Ratio (SNR) and Principal Component Analysis (PCA), etc. In this work, firstly, we propose to replace Points of Interests (POIs) and dimensionality reduction by utilizing word embedding, which converts power traces into sensitive vectors. Secondly, we combined sensitive vectors with Long Short Term Memories (LSTM) to execute SCA based on FPGA crypto-implementations. In addition, compared with traditional Template Attack (TA), Multiple Multilayer Perceptron (MLP) and Convolutional Neural Network (CNN). The result shows that the proposed model can not only reduce the manual operation, such as parametric assumptions and dimensionality setting, which limits their range of application, but improve the effectiveness of side-channel attacks as well.

Модели машинного обучения в задаче прогнозирования природно-ресурсного потенциала Пермского края

In the article we consider a complex indicator of region natural resource potential modeling and forecasting quality improvement using different machine learning models. Problem under consideration importance is determined by the fact that the models traditionally used for these purposes demonstrate either low quality, or high configuration and parameters evaluation difficulty. The aim of the study is determination of machine learning models that provide the optimal values of various modeling quality metrics. Materials and methods. For this study purposes we considered the multiple linear regression, decision tree, random forest, gradient boosting and multilayer perceptron mo¬dels. We used the determination coefficient R2, the root mean square error of modeling RMSE, the average absolute error of modeling MAE, and the relative error of prediction for 1 and 2 time intervals as quality metrics. This study is based on data of the complex indicator of the Perm Region natural resource potential and the system of its determining factors in the time interval from 2001 to 2018. We evaluate models and calculate quality metrics using Pandas and Scikit-learn Python libraries in Jupiter Notebook environment. Results. According to our research the classical multiple linear regression model demonstrates the worst results for all quality metrics under consideration. The decision tree model demonstrates determination coefficient maximum value and minimum root mean square error and mean absolute error. Minimum relative forecasting error for 2017 is provided by the gradient boosting model, for 2018 – by the multilayer perceptron model. Conclusion. Our study allows us to affirm that nonlinear machine learning models for the task of region natural resource potential modeling and forecasting demonstrate better approximating and predictive properties compared to multiple linear regression and thus can be used to improve the quality of natural resource management.

Utilization of multilayer perceptron for determining the inverse kinematics of an industrial robotic manipulator

Inverse kinematic equations allow the determination of the joint angles necessary for the robotic manipulator to place a tool into a predefined position. Determining this equation is vital but a complex work. In this article, an artificial neural network, more specifically, a feed-forward type, multilayer perceptron (MLP), is trained, so that it could be used to calculate the inverse kinematics for a robotic manipulator. First, direct kinematics of a robotic manipulator are determined using Denavit–Hartenberg method and a dataset of 15,000 points is generated using the calculated homogenous transformation matrices. Following that, multiple MLPs are trained with 10,240 different hyperparameter combinations to find the best. Each trained MLP is evaluated using the R 2 and mean absolute error metrics and the architectures of the MLPs that achieved the best results are presented. Results show a successful regression for the first five joints (percentage error being less than 0.1%) but a comparatively poor regression for the final joint due to the configuration of the robotic manipulator.

Machine learning tabulation of thermochemistry in turbulent combustion: An approach based on hybrid flamelet/random data and multiple multilayer perceptrons

A new machine learning methodology is proposed for speeding up thermochemistry computations in simulations of turbulent combustion. The approach is suited to a range of methods including Direct Numerical Simulation (DNS), Probability Density Function (PDF) methods, unsteady flamelet, Conditional Moment Closure (CMC), Multiple Mapping Closure (MMC), Linear Eddy Model (LEM), Thickened Flame Model, the Partially Stirred Reactor (PaSR) method (as in OpenFOAM) and the computation of laminar flames. In these methods, the chemical source term must be evaluated at every time step, and is often the most expensive element of a simulation. The proposed methodology has two main objectives: to offer enhanced capacity for generalisation and to improve the accuracy of the ANN prediction. To accomplish the first objective, we propose a hybrid flamelet/random data (HFRD) method for generating the training set. The random element endows the resulting ANNs with increased capacity for generalisation. Regarding the second objective, a multiple multilayer perceptron (MMP) approach is developed where different multilayer perceptrons (MLPs) are trained to predict states that result in smaller or larger composition changes, as these states feature different dynamics. It is shown that the multiple MLP method can greatly reduce the prediction error, especially for states yielding small composition changes. The approach is used to simulate flamelets of varying strain rates, one-dimensional premixed flames with differential diffusion and varying equivalence ratio, and finally the Large Eddy Simulation (LES) of CH4/air piloted flames Sandia D, E and F, which feature different levels of local extinction. The simulation results show very good agreement with those obtained from direct integration, while the range of problems simulated indicates that the approach has great capacity for generalisation. Finally, a speed-up ratio of 12 is attained for the reaction step.

Artificial intelligence-based nomogram for small-incision lenticule extraction

BackgroundSmall-incision lenticule extraction (SMILE) is a surgical procedure for the refractive correction of myopia and astigmatism, which has been reported as safe and effective. However, over- and under-correction still occur after SMILE. The necessity of nomograms is emphasized to achieve optimal refractive results. Ophthalmologists diagnose nomograms by analyzing the preoperative refractive data with their individual knowledge which they accumulate over years of experience. Our aim was to predict the nomograms of sphere, cylinder, and astigmatism axis for SMILE accurately by applying machine learning algorithm.MethodsWe retrospectively analyzed the data of 3,034 eyes composed of four categorical features and 28 numerical features selected from 46 features. The multiple linear regression, decision tree, AdaBoost, XGBoost, and multi-layer perceptron were employed in developing the nomogram models for sphere, cylinder, and astigmatism axis. The scores of the root-mean-square error (RMSE) and accuracy were evaluated and compared. Subsequently, the feature importance of the best models was calculated.ResultsAdaBoost achieved the highest performance with RMSE of 0.1378, 0.1166, and 5.17 for the sphere, cylinder, and astigmatism axis, respectively. The accuracies of which error below 0.25 D for the sphere and cylinder nomograms and 25° for the astigmatism axis nomograms were 0.969, 0.976, and 0.994, respectively. The feature with the highest importance was preoperative manifest refraction for all the cases of nomograms. For the sphere and cylinder nomograms, the following highly important feature was the surgeon.ConclusionsAmong the diverse machine learning algorithms, AdaBoost exhibited the highest performance in the prediction of the sphere, cylinder, and astigmatism axis nomograms for SMILE. The study proved the feasibility of applying artificial intelligence (AI) to nomograms for SMILE. Also, it may enhance the quality of the surgical result of SMILE by providing assistance in nomograms and preventing the misdiagnosis in nomograms.

CFFNN: Cross Feature Fusion Neural Network for Collaborative Filtering

Numerous state-of-the-art recommendation frameworks employ deep neural networks in Collaborative Filtering (CF). In this paper, we propose a cross feature fusion neural network (CFFNN) for the enhancement of CF. Existing studies overlook either user preferences for various item features or the relationship between item features and user features. To solve this problem, we construct a cross feature fusion network to enable the fusion of user features and item features as well as a self-attention network to determine users’ preferences for items. Specifically, we design a feature extraction layer with multiple MLP (Multilayer Perceptrons) modules to extract both user features and item features. Then, we introduce a cross feature fusion mechanism for an accurate determination of the relationship between different user-item interactions. The features of users and items are crossly embedded and then fed into a prediction network. The attention mechanism enables the model to focus on more effective features. The effectiveness of CFFNN model is demonstrated through extensive experiments on four real-world datasets. The experimental results indicate that CFFNN significantly outperforms the existing state-of-the-art models, with a relative improvement of 3.0 to 12.1 percent on hit ratio (HR) and normalized discounted cumulative gain (NDCG) compared with the baselines.

A multiple multilayer perceptron neural network with an adaptive learning algorithm for thyroid disease diagnosis in the internet of medical things

Medical information systems such as Internet of Medical Things (IoMT) are gained special attention over recent years. X-ray and MRI images are important sources of information to be examined for a particular type of anomalies. Reports based on the images and laboratory examination results could be mined with machine learning techniques as well. Thyroid disease diagnosis is an important capability of medical information systems. The main objective of this study is to improve the diagnosis accuracy of thyroid diseases from semantic reports and examination results using artificial neural network (ANN) in IoMT systems. In order to improve generalization and avoid over-fitting of ANN during the training process, a set of multiple multilayer perceptron (MMLP) neural network with the back-propagation error ability is proposed in this paper. Moreover, an adaptive learning rate algorithm is used to deal with the slow convergence and the local minima problem of the back-propagation error algorithm. The proposed MMLP significantly increased the overall accuracy of thyroid disease classification. With MMLP with a set of 6 networks, an improvement of 0.7% accuracy is achieved compared to a single network. In addition, comparing to the standard back-propagation, by using an adaptive learning rate algorithm in the proposed MMLP, an improvement of 4.6% accuracy and the final accuracy of 99% have been obtained in IoMT systems. The proposed MMLP is compared to recent researches reported for thyroid disease diagnosis, and its superiority is shown.