Feature Parameter Optimization Research Articles

Parameter optimization of the classifying device of straw micro-crusher based on neural network and CFD

To improve the classification performance of the straw micro-crusher's classifying device, the characteristic parameters impacting classification performance are identified through an analysis of the forces acting on the straw particles between the blades of the rotor cage and the motion characteristics of the fluid. Taguchi's experimental design is used to ascertain the combinations of computational fluid dynamics (CFD) simulation parameters for cut size and classifying sharpness index, while a neural network (NN) model is developed using CFD data to predict the optimal feature parameter combination for classification performance. The results reveal that the rotor cage speed holds the most significant impact on both the cut size and classifying sharpness index. The optimal combination of feature parameters recommended by the neural network is v=8m/s, n=1200r/min, z=36, θ=40°, at which the cut size is 28.3 μm, a reduction of 9.0 % compared to the Taguchi experiment. In addition, the relative error between the cut size predicted by the neural network model and the CFD simulation is less than 4.5 %, indicating the reliability of the neural network model in predicting the optimal parameter combination of the classifying device.

New method for optimizing the detection of low-energy soft X-rays (200–300 eV) in a CMOS detector

Complementary metal-oxide semiconductor (CMOS) sensors have been widely used as soft X-ray detectors in several fields owing to their recent developments and unique advantages. The CMOS detector parameters have been extensively studied and evaluated. However, the methods for discriminating signals and dark noise have not been sufficiently studied, particularly for low-energy soft X-rays. Setting a threshold, which is widely used for discrimination, is not suitable for low-energy soft X-rays because of the crossing of the energy bands of dark noise and signals. In this study, we analyzed the charge correlation of the CMOS detector GSENSE2020BSI (G2020BSI) and proposed a new two-dimensional segmentation method to optimize the discrimination of signals according to the charge correlation. The effectiveness of the method used to discriminate low-energy soft X-rays on the G2020BSI detector was qualitatively evaluated. The optimal feature parameters used in the two-dimensional segmentation method were investigated for G2020BSI. However, the two-dimensional segmentation method is insensitive to feature parameters.

An efficient drilling conditions classification method utilizing encoder and improved Graph Attention Networks

Identification of drilling conditions is paramount for ensuring safety and efficiency in drilling operations. Traditional methods, often based on manual analysis of drilling parameters and patterns by experts using empirical formulas, typically lack speed and precision in modern drilling procedures. To address this shortcoming, we have developed a sophisticated classification methodology that combines Encoder layers with improved Graph Attention Networks (GAT), aiming to precisely discern seven prevalent drilling conditions. Using an advanced expert empirical decision tree, we achieved superior accuracy in our dataset annotations. The model’s architecture integrates the K-nearest neighbors (KNN) algorithm for seamless integration between the Encoder layer and GAT networks. Sensitivity analyses pinpoint optimal feature parameters, leading to significant enhancements in model accuracy. Experimental outcomes show our annotation approach surpasses traditional techniques by 15.24%, achieving an accuracy of 96.73%. Additionally, our model boasts a recall rate of 94.22%, indicating high reliability in correctly identifying true positive drilling conditions, which is crucial for operational safety and decision-making in drilling scenarios. Characterized by its real-time capabilities, precision, and scalability, our model demonstrates promising potential for applications in practical drilling scenarios.

A single-building damage detection model based on multi-feature fusion: A case study in Yangbi

Accurate and effective identification, determination of the location, and classification of damaged buildings are essential after destructive earthquakes. However, the accuracy of image change detection is limited because of the many texture features and changes in non-building information. In this context, a model for single-building damage detection based on multi-feature fusion is proposed. First, the normalized Digital Surface Model (nDSM) was extracted from the DSM through iterative filtering and point cloud thinning, followed by the extraction of building contour information. Next, single-building images were generated from different data sources through the region of interest (ROI), and the optimal texture feature parameters were extracted for fusion. Afterward, principal-component analysis (PCA) was conducted to suppress multi-feature correlation-induced information redundancy. Finally, the damage to buildings was quantitatively evaluated, and the model was compared with 13 models. The results confirmed the practicability of the model for the Yangbi MS6.4 and Honghe MS5.0 earthquakes.

Prediction of extranodal extension in head and neck squamous cell carcinoma by CT images using an evolutionary learning model

BackgroundExtranodal extension (ENE) in head and neck squamous cell carcinoma (HNSCC) correlates to poor prognoses and influences treatment strategies. Deep learning may yield promising performance of predicting ENE in HNSCC but lack of transparency and interpretability. This work proposes an evolutionary learning method, called EL-ENE, to establish a more interpretable ENE prediction model for aiding clinical diagnosis.MethodsThere were 364 HNSCC patients who underwent neck lymph node (LN) dissection with pre-operative contrast-enhanced computerized tomography images. All the 778 LNs were divided into training and test sets with the ratio 8:2. EL-ENE uses an inheritable bi-objective combinatorial genetic algorithm for optimal feature selection and parameter setting of support vector machine. The diagnostic performances of the ENE prediction model and radiologists were compared using independent test datasets.ResultsThe EL-ENE model achieved the test accuracy of 80.00%, sensitivity of 81.13%, and specificity of 79.44% for ENE detection. The three radiologists achieved the mean diagnostic accuracy of 70.4%, sensitivity of 75.6%, and specificity of 67.9%. The features of gray-level texture and 3D morphology of LNs played essential roles in predicting ENE.ConclusionsThe EL-ENE method provided an accurate, comprehensible, and robust model to predict ENE in HNSCC with interpretable radiomic features for expanding clinical knowledge. The proposed transparent prediction models are more trustworthy and may increase their acceptance in daily clinical practice.

Diabetes risk prediction model based on community follow-up data using machine learning

Diabetes is a chronic metabolic disease characterized by hyperglycemia, the follow-up management of diabetes patients is mostly in the community, but the relationship between key lifestyle indicators in community follow-up and the risk of diabetes is unclear. In order to explore the association between key life characteristic indicators of community follow-up and the risk of diabetes, 252,176 follow-up records of people with diabetes patients from 2016 to 2023 were obtained from Haizhu District, Guangzhou. According to the follow-up data, the key life characteristic indicators that affect diabetes are determined, and the optimal feature subset is obtained through feature selection technology to accurately assess the risk of diabetes. A diabetes risk assessment model based on a random forest classifier was designed, which used optimal feature parameter selection and algorithm model comparison, with an accuracy of 91.24% and an AUC corresponding to the ROC curve of 97%. In order to improve the applicability of the model in clinical and real life, a diabetes risk score card was designed and tested using the original data, the accuracy was 95.15%, and the model reliability was high. The diabetes risk prediction model based on community follow-up big data mining can be used for large-scale risk screening and early warning by community doctors based on patient follow-up data, further promoting diabetes prevention and control strategies, and can also be used for wearable devices or intelligent biosensors for individual patient self examination, in order to improve lifestyle and reduce risk factor levels.

Modeling Ring Current Proton Fluxes Using Artificial Neural Network and Van Allen Probe Measurements

AbstractTerrestrial ring current dynamics are a critical part of the near‐space environment, in that they directly drive geomagnetic field variations that control particle drifts, and define geomagnetic storms. The present study aims to specify a global and time‐varying distribution of ring current proton using geomagnetic indices and solar wind parameters with their history as input. We train an artificial neural network (ANN) model to reproduce proton fluxes measured by the Radiation Belt Storm Probes Ion Composition Experiment instrument onboard Van Allen Probes. By choosing optimal feature parameters and their history length, the model results show a high correlation and a small error between model specifications and satellite measurements. The modeled results well capture energy‐dependent proton dynamics in association with geomagnetic storms, including inward radial diffusion, acceleration and decay. Our ANN model produces proton fluxes with their corresponding 3D spatiotemporal variations, capturing the latitudinal distribution and local time asymmetry that are consistent with observations and that can further inform theory.

A Gas Emission Prediction Model Based on Feature Selection and Improved Machine Learning

This paper proposed a gas emission prediction method based on feature selection and improved machine learning, as traditional gas emission prediction models are neither accurate nor universally applicable. Through analysis, this paper identified 12 factors that affected gas emissions. A total of 30 groups of typical data for gas outflow were standardized, after which a full subset regression feature selection method was used to categorize 12 influencing factors into different regular patterns and select 18 feature parameter sets. Meanwhile, based on nuclear principal component analysis (KPCA), an optimized gas emission prediction model was constructed where the dimensionality of the original data was reduced. An optimized algorithm set was constructed based on the hybrid kernel extreme learning machine (HKELM) and the least squares support vector machine (LSSVM). The performance of feature parameters adopted in the prediction algorithm was evaluated according to certain metrics. By comparing the results of different sets, the final prediction sequence could be obtained, and a model that was composed of the optimal feature parameters was applied to the optimal machine learning algorithm. The results showed that the HKELM outperformed LSSVM in prediction accuracy, running speed, and stability. The root meant square error (RMSE) for the final prediction sequence was 0.22865, the determination coefficient (R2) was 0.99395, the mean absolute error (MAE) was 0.20306, and the mean absolute percentage error (MAPE) was 1.0595%. Every index of accuracy evaluation performed well and the constructed prediction model had high-prediction accuracy and a wide application.

Speech-based gender recognition using linear prediction and mel-frequency cepstral coefficients

Gender discrimination and awareness are essentially practiced in social, education, workplace, and economic sectors across the globe. A person manifests this attribute naturally in gait, body gesture, facial, including speech. For that reason, automatic gender recognition (AGR) has become an interesting sub-topic in speech recognition systems that can be found in many speech technology applications. However, retrieving salient gender-related information from a speech signal is a challenging problem since speech contains abundant information apart from gender. The paper intends to compare the performance of human vocal tract-based model i.e., linear prediction coefficients (LPC) and human auditory-based model i.e., Mel-frequency cepstral coefficients (MFCC) which are popularly used in other speech recognition tasks by experimentation of optimal feature parameters and classifier’s parameters. The audio data used in this study was obtained from 93 speakers uttering selected words with different vowels. The two feature vectors were tested using two classification algorithms namely, discriminant analysis (DA) and artificial neural network (ANN). Although the experimental results were promising using both feature parameters, the best overall accuracy rate of 97.07% was recorded using MFCC-ANN techniques with almost equal performance for male and female classes.

Identify the triple-negative and non-triple-negative breast cancer by using texture features of medicale ultrasonic image: A STROBE-compliant study.

The study aimed to explore the value of ultrasound (US) texture analysis in the differential diagnosis of triple-negative breast cancer (TNBC) and non-TNBC.Retrospective analysis was done on 93 patients with breast cancer (35 patients with TNBC and 38 patients with non-TNBC) who were admitted to Taizhou people's hospital from July 2015 to June 2019. All lesions were pathologically proven at surgery. US images of all patients were collected. Texture analysis of US images was performed using MaZda software package. The differences between textural features in TNBC and non-TNBC were assessed. Receiver operating characteristic curve analysis was used to compare the diagnostic performance of textural parameters showing significant difference.Five optimal texture feature parameters were extracted from gray level run-length matrix, including gray level non-uniformity (GLNU) in horizontal direction, vertical gray level non-uniformity, GLNU in the 45 degree direction, run length non-uniformity in 135 degree direction, GLNU in the 135 degree direction. All these texture parameters were statistically higher in TNBC than in non-TNBC (P <.05). Receiver operating characteristic curve analysis indicated that at a threshold of 268.9068, GLNU in horizontal direction exhibited best diagnostic performance for differentiating TNBC from non-TNBC. Logistic regression model established based on all these parameters showed a sensitivity of 69.3%, specificity of 91.4% and area under the curve of 0.834.US texture features were significantly different between TNBC and non-TNBC, US texture analysis can be used for preliminary differentiation of TNBC from non-TNBC.

An identification method of cashmere and wool by the two features fusion

PurposeThe purpose of this paper is to select the optimal feature parameters to further improve the identification accuracy of cashmere and wool.Design/methodology/approachTo increase the accuracy, the authors put forward a method selecting optimal parameters based on the fusion of morphological feature and texture feature. The first step is to acquire the fiber diameter measured by the central axis algorithm. The second step is to acquire the optimal texture feature parameters. This step is mainly achieved by using the variance of secondary statistics of these two texture features to get four statistics and then finding the impact factors of gray level co-occurrence matrix relying on the relationship between the secondary statistic values and the pixel pitch. Finally, the five-dimensional feature vectors extracted from the sample image are fed into the fisher classifier.FindingsThe improvement of identification accuracy can be achieved by determining the optimal feature parameters and fusing two texture features. The average identification accuracy is 96.713% in this paper, which is very helpful to improve the efficiency of detector in the textile industry.Originality/valueIn this paper, a novel identification method which extracts the optimal feature parameter is proposed.

Monitoring of rice lodging using Sentinel-1 data

Synthetic Aperture Radar (SAR) can monitor rice regardless of time and weather condition, and the ability of crop lodge monitoring has been demonstrated by previous studies. However, there have no study about monitoring of rice lodging using satellite-based SAR. In this study, we extracted backscatter coefficient (BC) and H/A/Alpha polarimetric parameters from dual-polarized Sentinel-1 SAR data, and a lodging sensitivity factor γ was constructed for selecting optimal radar feature parameter (ORFP) which is highly sensitivity to lodging rice. Then, the decision tree classification method was used with multiple ORFPs to classify the healthy and lodging rice plots. We firstly acquired shape of rice region using Sentinel-2 image and Maximum likelihood classification method for eliminating the influence of other features before classification of lodging rice. The result showed that the overall accuracy of 87.18% is achieved with the combination of ORFPs and decision tree.

LDNFSGB: prediction of long non-coding rna and disease association using network feature similarity and gradient boosting

BackgroundA large number of experimental studies show that the mutation and regulation of long non-coding RNAs (lncRNAs) are associated with various human diseases. Accurate prediction of lncRNA-disease associations can provide a new perspective for the diagnosis and treatment of diseases. The main function of many lncRNAs is still unclear and using traditional experiments to detect lncRNA-disease associations is time-consuming.ResultsIn this paper, we develop a novel and effective method for the prediction of lncRNA-disease associations using network feature similarity and gradient boosting (LDNFSGB). In LDNFSGB, we first construct a comprehensive feature vector to effectively extract the global and local information of lncRNAs and diseases through considering the disease semantic similarity (DISSS), the lncRNA function similarity (LNCFS), the lncRNA Gaussian interaction profile kernel similarity (LNCGS), the disease Gaussian interaction profile kernel similarity (DISGS), and the lncRNA-disease interaction (LNCDIS). Particularly, two methods are used to calculate the DISSS (LNCFS) for considering the local and global information of disease semantics (lncRNA functions) respectively. An autoencoder is then used to reduce the dimensionality of the feature vector to obtain the optimal feature parameter from the original feature set. Furthermore, we employ the gradient boosting algorithm to obtain the lncRNA-disease association prediction.ConclusionsIn this study, hold-out, leave-one-out cross-validation, and ten-fold cross-validation methods are implemented on three publicly available datasets to evaluate the performance of LDNFSGB. Extensive experiments show that LDNFSGB dramatically outperforms other state-of-the-art methods. The case studies on six diseases, including cancers and non-cancers, further demonstrate the effectiveness of our method in real-world applications.

A tool condition monitoring method based on two-layer angle kernel extreme learning machine and binary differential evolution for milling

Tool condition monitoring (TCM) in numerical control (NC) machines plays a significant role in ensuring manufacturing quality. TCM based on multiple sensors can provide more information related to tool condition and is a topic of great interest. A novel TCM method for milling processes based on a two-layer angle kernel extreme learning machine (TAKELM) and binary differential evolution (BDE) was proposed. The TAKELM could enhance inherent feature extraction in microarray data and don’t require manual presetting of the kernel function or optimization of its hyperparameter. The BDE algorithm was applied to search the optimal feature parameter combinations in multi-domain alternative feature parameters to achieve the fewest feature parameters satisfying a very small prediction error. Finally, the performance of the proposed TAKELM- BDE method was verified on two milling TCM experiments (one open-access benchmark TCM data set and one TCM experiment) by comparison to the KELM- BDE, PCC- TAKELM, and mRMR- TAKELM methods. The results indicated that the mean absolute error, mean absolute percentage error, and root mean square error of the proposed method are less than 0.005 on two milling TCM experiments, and outperforms the other three methods.

Strengthen EEG-based emotion recognition using firefly integrated optimization algorithm

Emotion recognition is helpful for human to enhance self-awareness and respond appropriately towards the happenings around them. Due to the complexity and diversity of emotions, EEG-based emotion recognition is still a challenging task in pattern recognition. In order to recognize diverse emotions, we propose a novel firefly integrated optimization algorithm (FIOA) in this paper. It can simultaneously accomplish multiple tasks, i.e. the optimal feature selection, parameter setting and classifier selection according to different EEG-based emotion datasets. The FIOA utilizes a ranking probability objection function to guarantee the high accuracy recognition with less features. Moreover, the hybrid encoding expression and the dual updating strategy are developed in the FIOA so as to realize the optimal selection of feature subset and classifier without stagnating in the local optimum. In addition to the public DEAP datasets, we also conducted an EEG-based music emotion experiment involving 20 subjects for the validation of the proposed FIOA. After filtering and segmentation, three categories of features were extracted from every EEG signal. Then FIOA was applied to every subject dataset for two pattern recognition of emotions. The results show that the FIOA can automatically find the optimal features, parameter and classifier for different emotion datasets, which greatly reduces the artificial selection workload. Furthermore, comparing with the binary particle swarm optimization (PSObinary) and the binary firefly (FAbinary), the FIOA can achieve the higher accuracy with less features in the emotion recognition.

An Optimal Feature Parameter Set Based on Gated Recurrent Unit Recurrent Neural Networks for Speech Segment Detection

Speech segment detection based on gated recurrent unit (GRU) recurrent neural networks for the Kurdish language was investigated in the present study. The novelties of the current research are the utilization of a GRU in Kurdish speech segment detection, creation of a unique database from the Kurdish language, and optimization of processing parameters for Kurdish speech segmentation. This study is the first attempt to find the optimal feature parameters of the model and to form a large Kurdish vocabulary dataset for a speech segment detection based on consonant, vowel, and silence (C/V/S) discrimination. For this purpose, four window sizes and three window types with three hybrid feature vector techniques were used to describe the phoneme boundaries. Identification of the phoneme boundaries using a GRU recurrent neural network was performed with six different classification algorithms for the C/V/S discrimination. We have demonstrated that the GRU model has achieved outstanding speech segmentation performance for characterizing Kurdish acoustic signals. The experimental findings of the present study show the significance of the segment detection of speech signals by effectively utilizing hybrid features, window sizes, window types, and classification models for Kurdish speech.

Landform classification based on optimal texture feature extraction from DEM data in Shandong Hilly Area, China

Texture and its analysis methods are crucial for image feature extraction and classification. Digital elevation model (DEM) is the most important data source of digital terrain analysis and landform classification, and considerable research values are gained from texture feature extraction and analysis from DEM data. In this research, on the basis of optimal texture feature extraction, the hilly area in Shandong, China, was selected as the study area, and DEM data with a resolution of 500 m were used as the experimental data for landform classification. First, second-order texture measures and texture image were extracted from DEM data by using a gray level cooccurrence matrix (GLCM). Second, the variation characteristics of each texture measure were analyzed, and the optimal feature parameters, such as direction, gray level, and texture window, were determined. Meanwhile, the texture feature value, combined with maximum information, was calculated, and the multiband texture image was obtained by resolving three optimal texture measure images. Finally, a support vector machine (SVM) method was adopted to classify landforms on the basis of the multiband texture image. Results indicated that the texture features of DEM data can be sufficiently represented and measured via the quantitative GLCM method. However, the feature parameters during the texture feature value calculation required further optimization. Based on the image texture from DEM data, efficient classification accuracy and ideal classification effect were achieved.

Learning Compact Model for Large-Scale Multi-Label Data

Large-scale multi-label learning (LMLL) aims to annotate relevant labels from a large number of candidates for unseen data. Due to the high dimensionality in both feature and label spaces in LMLL, the storage overheads of LMLL models are often costly. This paper proposes a POP (joint label and feature Parameter OPtimization) method. It tries to filter out redundant model parameters to facilitate compact models. Our key insights are as follows. First, we investigate labels that have little impact on the commonly used LMLL performance metrics and only preserve a small number of dominant parameters for these labels. Second, for the remaining influential labels, we reduce spurious feature parameters that have little contribution to the generalization capability of models, and preserve parameters for only discriminative features. The overall problem is formulated as a constrained optimization problem pursuing minimal model size. In order to solve the resultant difficult optimization, we show that a relaxation of the optimization can be efficiently solved using binary search and greedy strategies. Experiments verify that the proposed method clearly reduces the model size compared to state-of-the-art LMLL approaches, in addition, achieves highly competitive performance.

Improved Artificial Bee Colony Algorithm and its Application in Classification

For improving the classification accuracy of the classifier, a novel classification methodology based on artificial bee colony algorithm is proposed for optimal feature and SVM parameters selection. In order to balance the ability of exploration and exploitation of traditional ABC algorithm, improvements are introduced for the generation of initial solution set and onlooker bee stage. The proposed algorithm is applied to four datasets with different attribute characteristics from UCI and efficiency of the algorithm is proved from the results.

Detection of non-small cell lung cancer cells based on microfluidic polarization microscopic image analysis.

In early diagnosis of lung cancer, a polarization microscopy is a powerful tool to obtain the optical information of biological tissues. In this paper, a new microfluidic polarization imaging and analysis method was proposed for the detection and classification of cancer-associated fibroblasts and the two kinds of non-small cell lung cancer cells, A549 and H322. A polarizing microscopy system was constructed based on a commercial microscope to obtain 3*3 Mueller matrix of cells. Based on the Muller matrix decomposition algorithm and analysis in spatial domain and frequency domain, appropriate classification parameters were selected for the characterization of different polarization characteristics of cells. Finally, the logistic regression models based on machine learning were applied to determine optimal feature parameters and classify cells. This method integrated the morphological information of the cells, and the polarization characteristics of the cells in different polarization states. It is for the first time that the polarization microscopic image analysis method has been applied to the detection and classification of non-small cell lung cancer cells. The results show that the presented microfluidic polarization microscopic image analysis method could classify cells effectively. Compared with the Muller matrix measurement and calculation methods, the method proposed in this paper was greatly simplified in both the acquisition of polarized images and the analysis and processing of polarized images.