Railway dangerous goods transportation system risk identification: Comparisons among SVM, PSO-SVM, GA-SVM and GS-SVM

Study on the Determination of Flavor Value of Rice Based on Grid Iterative Search Swarm Optimization Support Vector Machine Model and Hyperspectral Imaging

In the field of rice processing and cultivation, it is crucial to adopt efficient, rapid and user-friendly techniques to detect the flavor values of various rice varieties. The conventional methods for flavor value assessment mainly rely on chemical analysis and technical evaluation, which not only deplete the rice resources but also incur significant time and labor costs. In this study, hyperspectral imaging technology was utilized in combination with an improved Particle Swarm Optimization Support Vector Machine (PSO-SVM) algorithm, i.e., the Grid Iterative Search Particle Swarm Optimization Support Vector Machine (GISPSO-SVM) algorithm, introducing a new non-destructive technique to determine the flavor value of rice. The method captures the hyperspectral feature data of different rice varieties through image acquisition, preprocessing and feature extraction, and then uses these features to train a model using an optimized machine learning algorithm. The results show that the introduction of GIS algorithms in a PSO-optimized SVM is very effective and can improve the parameter finding ability. In terms of flavor value prediction accuracy, the Principal Component Analysis (PCA) combined with the GISPSO-SVM algorithm achieved 96% accuracy, which was higher than the 93% of the Competitive Adaptive Weighted Sampling (CARS) algorithm. And the introduction of the GIS algorithm in different feature selection can improve the accuracy to different degrees. This novel approach helps to evaluate the flavor values of new rice varieties non-destructively and provides a new perspective for future rice flavor value detection methods.

Raman spectroscopy with an improved support vector machine for discrimination of thyroid and parathyroid tissues

The objective of this study was to discriminate thyroid and parathyroid tissues using Raman spectroscopy combined with an improved support vector machine (SVM) algorithm. In thyroid surgery, there is a risk of inadvertently removing the parathyroid glands. At present, there is a lack of research on using Raman spectroscopy to discriminate parathyroid and thyroid tissues. In this article, samples were obtained from 43 individuals with thyroid and parathyroid tissues for Raman spectroscopy analysis. This study employed partial least squares (PLS) to reduce dimensions of data, and three optimization algorithms are used to improve the classification accuracy of SVM algorithm model in spectral analysis. The results show that PLS-GA-SVM algorithm has higher diagnostic accuracy and better reliability. The sensitivity of this algorithm is 94.67% and the accuracy is 94.44%. It can be concluded that Raman spectroscopy combined with the PLS-GA-SVM diagnostic algorithm has significant potential for discriminating thyroid and parathyroid tissues.

Delicately Reinforced k-Nearest Neighbor Classifier Combined With Expert Knowledge Applied to Abnormity Forecast in Electrolytic Cell

As the profit and safety requirements become higher and higher, it is more and more necessary to realize an advanced intelligent analysis for abnormity forecast of the synthetical balance of material and energy (AF-SBME) on aluminum reduction cells (ARCs). Without loss of generality, AF-SBME belongs to classification problems. Its advanced intelligent analysis can be realized by high-performance data-driven classifiers. However, AF-SBME has some difficulties, including a high requirement for interpretability of data-driven classifiers, a small number, and decreasing-over-time correctness of training samples. In this article, based on a preferable data-driven classifier, which is called a reinforced k -nearest neighbor (R-KNN) classifier, a delicately R-KNN combined with expert knowledge (DR-KNN/CE) is proposed. It improves R-KNN in two ways, including using expert knowledge as external assistance and enhancing self-ability to mine and synthesize data knowledge. The related experiments on AF-SBME, where the relevant data are directly sampled from practical production, have demonstrated that the proposed DR-KNN/CE not only makes an effective improvement for R-KNN, but also has a more advanced performance compared with other existing high-performance data-driven classifiers.

Mining the accident causes of railway dangerous goods transportation: A Logistics-DT-TFP based approach

Accurately and quickly mining the hidden information in railway dangerous goods transportation (RDGT) accident reports has great significance for its safety management. In this paper, a data mining method Logistics-DT-TFP is proposed for analysing the causes of RDGT accidents. Firstly, analyse the transportation process, extract the cause of the accident, and classify the severity of the accident. Then, using ordered multi-classification Logistic regression for correlation calculation, qualitatively judge and quantitatively analyse the relationship between each cause and the severity of the accident. The feature tags of the Decision Tree (DT) are screened, the C5.0 algorithm is used to obtain the accident coupling rules. Next, the FP-Growth algorithm is used to mine frequent itemsets, and TOP-K is used to improve it and output effective association rules with the degree of lift as the indicator, which avoids repeated traversal of the database, shortens the time complexity, and reduces the impact of the minimum support setting on the calculation results. The degree of lift among the causes in the coupling chain is calculated as a complement to the extraction of coupling rules. Finally, based on the analysis and mining results of case study, the management strategies for railway dangerous goods are proposed.

Mechanomyography signals pattern recognition in hand movements using swarm intelligence algorithm optimized support vector machine based on acceleration sensors

On the basis of extracting mechanomyography (MMG) signal features, the classification of hand movements has certain application values in human-machine interaction systems and wearable devices. In this paper, pattern recognition of hand movements based on MMG signal is studied with swarm intelligence algorithms introduced to optimize support vector machine (SVM). Time domain (TD) features, wavelet packet node energy (WPNE) features, frequency domain (FD) features, convolution neural network (CNN) features were extracted from each channel to constitute different feature sets. Three novel swarm intelligence algorithms (i.e., bald eagle search (BES), sparrow search algorithm (SSA), grey wolf optimization (GWO)) optimized SVM is proposed to train the models and recognition of hand movements are tested for each MMG feature extraction method. Using GWO as the optimization algorithm, time consumption is less than using the other two swarm algorithms. Using GWO with TD+FD features can obtain the classification accuracy of 93.55 %, which is higher than other methods while using CNN to extract features can be independent of domain knowledge. The results confirm GWO-SVM with TD + FD features is superior to some other methods in the classification problem for tiny samples based on MMG.

A detection method of typical toxic mixed red tide algae in Qinhuangdao based on three-dimensional fluorescence spectroscopy

Red tides occur every year in the Qinhuangdao sea area of China, including a variety of toxic algae and non-toxic algae. Toxic red tide algae have caused great damage to the marine aquaculture industry in China and seriously endangered human health, but most of non-toxic algae are important bait for marine plankton. Therefore, it is very important to identify the type of mixed red tide algae in Qinhuangdao sea area. In this paper, three-dimensional fluorescence spectroscopy and chemometrics were applied to the identification of typical toxic mixed red tide algae in Qinhuangdao. Firstly, the three-dimensional fluorescence spectrum data of typical mixed red tide algae in Qinhuangdao sea area were measured by f-7000 fluorescence spectrometer, and the contour map of algae samples was obtained. Secondly, the contour spectrum analysis is carried out to find the excitation wavelength of the peak position of the three-dimensional fluorescence spectrum and form the new three-dimensional fluorescence spectrum data selected by the feature interval. Then, the new three-dimensional fluorescence spectrum data are extracted by principal component analysis (PCA). Finally, the feature extraction data and the data without feature extraction are used as the input of the genetic optimization support vector machine (GA-SVM) and particle swarm optimization support vector machine (PSO-SVM) classification models, respectively, to obtain the classification model of mixed red tide algae, and the two feature extraction analysis methods and two classification algorithms are compared. The results show that the classification accuracy of the test set using the principal component feature extraction and GA-SVM classification method is 92.97 %, when the excitation wavelengths are 420 nm, 440 nm, 480 nm, 500 nm and 580 nm, and the emission wavelengths are 650-750 nm. Therefore, it is feasible and effective to apply the three-dimensional fluorescence spectrum characteristics and genetic optimization support vector machine classification method to the identification of toxic mixed red tide algae in Qinhuangdao sea area.

Impact of green technology innovation based on IoT and industrial supply chain on the promotion of enterprise digital economy

With the gradual deterioration of the natural environment, a green economy has become a competing goal for all countries. As a trend of green innovation development, the digital economy has become a research hotspot for scientists. In this article, we study the supply chain management of enterprises in green innovation and digital economy development and complete the identification and demand prediction of warehouse goods through the Internet of Things (IoT) and artificial intelligence (AI). As the stuff meets the goods detection and storage, we employ an intelligent method to detect and classify the goods. The demand prediction analysis is carried out based on historical data on goods demand in the enterprise. The absolute error between the prediction result and the actual demand within 1 week is less than 30 goods by the particle swarm optimization-support vector machine (PSO-SVM) method used in this article. First, the goods identification task is completed based on video surveillance data using YOLOv4, and the recognition rate is as high as 98.3%. This article realises enterprises' intelligent supply chain management through the intelligent identification of goods and the demand forecasting analysis of goods in the warehouse, which provides new ideas for green innovation and digital economy development.

Detection and classification of cracks and potholes in road images using texture descriptors

Traffic safety is directly affected by poor road conditions. Automating the detection of road defects allows improvements in the maintenance process. The identification of defects such as cracks and potholes can be done using computer vision techniques and supervised learning. In this paper, we propose the detection of cracks and potholes in images of paved roads using machine learning techniques. The images are subdivided into blocks, where Gray-Level Co-Occurrence Matrix (GLCM), Local Binary Pattern (LBP), and Gabor Filter’s texture descriptors are used to extract features of the images. For the classification task, the Support Vector Machines (SVM), k-Nearest Neighbors (kNN), and Multi-Layer Perceptron (MLP) models are compared. We performed two experiments on a dataset built with images of Brazilian highways. In the first experiment, we obtained a F-measure of 75.16% when classifying blocks of images that have cracks and potholes, and 79.56% when comparing roads with defects and without defects. In the second experiment, a F-measure of 87.06% was obtained for the equivalent task. Thus, it is possible to state that the use of the techniques presented is feasible for locating faults in highways.

Structural Analysis and Classification of Low-Molecular-Weight Hyaluronic Acid by Near-Infrared Spectroscopy: A Comparison between Traditional Machine Learning and Deep Learning

Confusing low-molecular-weight hyaluronic acid (LMWHA) from acid degradation and enzymatic hydrolysis (named LMWHA-A and LMWHA-E, respectively) will lead to health hazards and commercial risks. The purpose of this work is to analyze the structural differences between LMWHA-A and LMWHA-E, and then achieve a fast and accurate classification based on near-infrared (NIR) spectroscopy and machine learning. First, we combined nuclear magnetic resonance (NMR), Fourier transform infrared (FTIR) spectroscopy, two-dimensional correlated NIR spectroscopy (2DCOS), and aquaphotomics to analyze the structural differences between LMWHA-A and LMWHA-E. Second, we compared the dimensionality reduction methods including principal component analysis (PCA), kernel PCA (KPCA), and t-distributed stochastic neighbor embedding (t-SNE). Finally, the differences in classification effect of traditional machine learning methods including partial least squares-discriminant analysis (PLS-DA), support vector classification (SVC), and random forest (RF) as well as deep learning methods including one-dimensional convolutional neural network (1D-CNN) and long short-term memory (LSTM) were compared. The results showed that genetic algorithm (GA)-SVC and RF were the best performers in traditional machine learning, but their highest accuracy in the test dataset was 90%, while the accuracy of 1D-CNN and LSTM models in the training dataset and test dataset classification was 100%. The results of this study show that compared with traditional machine learning, the deep learning models were better for the classification of LMWHA-A and LMWHA-E. Our research provides a new methodological reference for the rapid and accurate classification of biological macromolecules.

Rapid Identification of Soybean Varieties by Terahertz Frequency-Domain Spectroscopy and Grey Wolf Optimizer-Support Vector Machine

Different soybean varieties vary greatly in their nutritional value and composition. Screening for superior varieties is also essential for the development of the soybean seed industry. The objective of the paper was to analyze the feasibility of terahertz (THz) frequency-domain spectroscopy and chemometrics for soybean variety identification. Meanwhile, a grey wolf optimizer-support vector machine (GWO-SVM) soybean variety identification model was proposed. Firstly, the THz frequency-domain spectra of experimental samples (6 varieties, 270 in total) were collected. Principal component analysis (PCA) was used to analyze the THz spectra. After that, 203 samples from the calibration set were used to establish a soybean variety identification model. Finally, 67 samples from the test set were used for prediction validation. The experimental results demonstrated that THz frequency-domain spectroscopy combined with GWO-SVM could quickly and accurately identify soybean varieties. Compared with discriminant partial least squares (DPLS) and particles swarm optimization support vector machine, GWO-SVM combined with the second derivative could establish a better soybean variety identification model. The overall correct identification rate of its prediction set was 97.01%.

Interpretable Machine Learning Models for Punching Shear Strength Estimation of FRP Reinforced Concrete Slabs

Fiber reinforced polymer (FRP) serves as a prospective alternative to reinforcement in concrete slabs. However, similarly to traditional reinforced concrete slabs, FRP reinforced concrete slabs are susceptible to punching shear failure. Accounts of the insufficient consideration of impact factors, existing empirical models and design provisions for punching strength of FRP reinforced concrete slabs have some problems such as high bias and variance. This study established machine learning-based models to accurately predict the punching shear strength of FRP reinforced concrete slabs. A database of 121 groups of experimental results of FRP reinforced concrete slabs are collected from a literature review. Several machine learning algorithms, such as artificial neural network, support vector machine, decision tree, and adaptive boosting, are selected to build models and compare the performance between them. To demonstrate the predicted accuracy of machine learning, this paper also introduces 6 empirical models and design codes for comparative analysis. The comparative results demonstrate that adaptive boosting has the highest predicted precision, in which the root mean squared error, mean absolute error and coefficient of determination of which are 29.83, 23.00 and 0.99, respectively. GB 50010-2010 (2015) has the best predicted performance among these empirical models and design codes, and ACI 318-19 has the similar result. In addition, among these empirical models, the model proposed by El-Ghandour et al. (1999) has the highest predicted accuracy. According to the results obtained above, SHapley Additive exPlanation (SHAP) is adopted to illustrate the predicted process of AdaBoost. SHAP not only provides global and individual interpretations, but also carries out feature dependency analysis for each input variable. The interpretation results of the model reflect the importance and contribution of the factors that influence the punching shear strength in the machine learning model.