Single Algorithm Research Articles

Influence of stationary porous media on fluid flow

The work examines the flow of an incompressible fluid containing a porous medium. A porous body is formed through various mechanisms. To describe the flow, a single equation is used that describes the flow of fluid in the porous and free zone. The flow is simulated based on Rakhmatulin’s two-speed model, in a laminar mode with zero speed of a discrete phase. The results of numerical simulation of the hydrodynamic features of a two-dimensional viscous flow are presented. The Kozeny-Karman relation is used as the force of interaction with the porous layer. Computational experimental methods are used to study the effects of nonuniformity of the fluid velocity field arising from a porous body. For the numerical implementation of the resulting equation, which is a generalization of the Navier-Stokes equation, a SIMPLE-like algorithm with corresponding generalizations was used. A single algorithm is used for the entire area, without identifying the free and porous zones.

Prediction of the Apple Sugar Content of Multiple Varieties by Dielectric Spectroscopy

Highlights Explore two distinct types of frequency features: optimal frequency and full frequency. Utilize a one-dimensional convolutional neural network (1D-CNN) to classify varieties of apples, employing dielectric spectroscopy features as inputs. Establish models for predicting apple sugar content in various hybrids. Develop an optimal multi-variety search model by integrating a variety recognition model, an individual-variety prediction model, and a classification-first search strategy for accurate soluble solids content (SSC) prediction. Abstract. As a non-destructive detection technology, dielectric spectroscopy encounters challenges when attempting to detect apple sugar content, particularly when aiming to predict sugar content accurately. Variety-specific models may be necessary due to the distinct composition and cell structures present in different apple varieties. Consequently, a single algorithm may prove insufficient for predicting sugar content with precision. Therefore, it becomes imperative to achieve precise sugar content detection across various apple varieties. In the present study, we conducted an analysis of spectral data obtained from apple samples through dielectric spectroscopy. Leveraging these dielectric spectra as input variables, we devised a multi-variety model for predicting apple sugar content. Within this model, the successive projections algorithm (SPA) method was employed to select and optimize the frequency variables. Furthermore, a one-dimensional convolutional neural network (1D-CNN) was applied to the established variety classification model, and the PLS method was employed to develop the individual-variety and hybrid apple sugar content prediction models. When compared to the other two models, the multi-variety model demonstrated superior accuracy in predicting sugar content across various apple varieties. The R2 p and RMSEP overall prediction sets were 0.915 and 0.00147° Brix, respectively. Meanwhile, the lower values of RMSEP in the other three varieties were 0.0015, 0.0012, and 0.0017° Brix, respectively. This finding indicated that the proposed multi-variety search model was sensitive to variety differences in the prediction of apple sugar content. Therefore, it could improve the accuracy and stability of apple sugar prediction. Keywords: Apple sugar content, Deep learning, Dielectric spectroscopy, Multi-variety search model.

ПОКРАЩЕННЯ ВІДЕОПОСЛІДОВНОСТІ В СИСТЕМАХ ВІДЕОАНАЛІТИКИ

A video enhancement method based on applying gamma correction is proposed. Its realization implies processing all video frames using a single algorithm. The proposed algorithm, due to the preservation of inter-frame coherence throughout the entire video, significantly reduces the time spent searching for the optimal value of the gamma parameter, which ensures the highest quality of the frame reaching the maximum possible level in the automatic mode in the sense of both visual observation and the determination of key points and the selection of contours of objects in images. The method is characterized by high adaptability to sudden changes in the lighting of the scene, preservation of interframe coherence, and the absence of any side negative artifacts in the enhanced video. A toolkit for automatically determining the optimal value of the gamma parameter for video frames is developed. It significantly increases the efficiency of video analytics systems, image and video segmentation and processing processes due to reducing the negative impact of the scene lighting mode on image quality. Keywords: gamma-correction, video analytics system, video sequence enhancement, histogram, cumulative histogram, video processing, interframe coherence.

A new hybrid classification algorithm for predicting student performance

<abstract><p>Education is essential and increasingly crucial for the development of almost all countries worldwide. As educational data has become increasingly available, scholars have shown a growing interest in exploring the correlation between students' academic achievements and other factors that may impact their performance using machine learning algorithms. This research paper introduces a novel hybrid classifier that aims to predict the academic performance of students by using a combination of different single algorithms. The proposed hybrid classifier (PHC) is compared to six available classification algorithms (random forest (RF), C4.5, classification and regression trees (CART), support vector machines (SVM), naive Bayes (NB) and K-nearest neighbors (KNN)) using recall, precision, F1-score, and accuracy evaluation measures. Our experimental results reveal that the PHC classifier consistently outperforms the individual classifiers across multiple evaluation metrics. Specifically, the PHC classifier achieved an accuracy rate of 92.40%, surpassing the RF, C4.5, and CART classifiers, which were the next best performers. In terms of precision and F1 score, the PHC also demonstrated superior performance, indicating its robustness in correctly identifying positive instances and providing balanced accuracy. While the C4.5 classifier performed comparably to the PHC classifier concerning the recall metric, the hybrid model's overall performance highlights its effectiveness in leveraging the complementary strengths of the included classifiers. The suggested hybrid model has the potential to enhance students' academic performance and success more effectively and efficiently. It could benefit students, educators, and academic institutions. Additionally, it provides practical insights for educators and institutions striving to improve student achievement using predictive analysis.</p></abstract>

A Neuro Phenotypic Evolution Algorithm for Recognizing Human Motion Type

Living in the modern world requires having a precise, intelligent system that may be suggested to do various activities. Due to the scalability of AI algorithms, we have proposed a phenotypic Evolutionary Algorithm (EA)-based system to assist the Artificial Neural Network algorithm (ANN) in the learning process. Combining the two strategies can result in a smart neuro-evolutionary model that is effective in accomplishing significant duties in various domains. The suggested multi-layer neural algorithm's design creates the conditions for learning via the EA’s processes of crossover, mutation, and selection. To aid in the selection and crossover processes, the learning process phase breaks up the original ANN into multiple ANNs according to the number of hidden layers. ANNs are ranked from worst to best in the selection phase based on the soring function that is applied to the fitness list. The fitness list retains each ANN’s accuracy even after breaking apart the original ANN. The crossover procedure is then applied between the two worst and best ANNs. Mutation provides a means of improvement for the less effective ANNs. Following completion of these processes, the ANN algorithms are combined to create a single ANN algorithm. The Vicon mobile robot (SCITOS G5) system’s multi-dimensional data, which extracted both aggressive and typical human movement, as well as Human Activity Recognition (HAR) datasets extracted by smartphones, both have been applied using the suggested method. The system achieved a high performance and efficiency rate on the intended recognition problem

Operator-free HPLC automated method development guided by Bayesian optimization

Automated, closed-loop HPLC method optimization using single and multi-objective Bayesian optimization algorithms.

Research on Path Planning for Unmanned Surface Vessels Based on AIS Data

To address the issues of poor endpoint convergence and suboptimal path quality in global path planning for unmanned surface vessels (USVs), this paper proposes an endpoint-convergence oriented improved genetic algorithm based on AIS data. Firstly, utilizing a genetic algorithm with an endpoint-convergence objective function, a set of paths with strong endpoint convergence is selected. Secondly, to overcome the limitations of a single optimization algorithm, a multi-objective path planning approach is employed using a genetic algorithm, thereby improving the distance and trajectory smoothness in global path planning for unmanned ships. Finally, the planned paths are evaluated by comparing them with the real paths from AIS data. The simulation results demonstrate that the proposed method outperforms other traditional algorithms in terms of average path turn count and turn angle, reducing them by an average of 35.41% and 35.72%, respectively. Moreover, the proposed method exhibits the smallest error compared to the real trajectories, with an average reduction of 18.26%. These results validate the effectiveness and rationality of the proposed approach in improving path quality, reducing path turn count, and achieving better alignment with actual trajectories.

Monte Carlo global variance reduction method combining source bias and weight window

The global tally problem has a wide range of applications in major research fields such as Monte Carlo simulations of pin-by-pin reactor models and time-dependent particle transport problems in multi-physics coupling calculations. Due to the uneven power distribution of the simulated system, the statistical errors of all tallies are unevenly distributed, resulting in some low global efficiency. For this kind of problem with global characteristics, it is necessary to develop global variance reduction techniques to obtain the accurate distribution of target tallies in the entire space. A large number of global variance reduction algorithms have been studied based on the consideration of flattening global tally error distribution, so as to improve global efficiency. This work focuses on the combination of two efficient global variance reduction algorithms, namely, the uniform tally density algorithm and the weight window algorithm, which belong to source bias and transport process bias, respectively. In tally, a method is proposed to adjust the weight window parameters by using the bias factor of the uniform tally density algorithm. Then, the weight window method will be used to reduce the weight fluctuation caused by the uniform tally density method. In this way, an organic combination of these two algorithms can be realized. A series of comparative tests are carried out based on the Hoogenboom-Martin pressurized water reactor benchmark, and it is verified that the hybrid global variance reduction algorithm proposed in this work is better than the single weight window algorithm or the uniform tally density algorithm. In terms of reducing the maximum error, the global efficiency of the hybrid algorithm is 2.6 times and 3 times that of the weight window algorithm and the uniform tally density algorithm, respectively. In addition, through the comparative analysis of computational asymmetry degree and computational efficiency, it is verified that the uniform tally density algorithm has better performance than the classical uniform fission site algorithm, and the performance advantages of the uniform tally density algorithm are quantitatively evaluated based on some new indicators. The results show that the hybrid global variance reduction algorithm proposed in this work can solve the global tally problem efficiently, thereby further promoting research in related fields.

Prognosis and therapy in thyroid cancer by gene signatures related to natural killer cells.

Natural killer (NK) cells are crucial to cancer development and prognosis. However, the role of NK cell-related genes in immunotherapy and the tumor immune microenvironment (TIME) is not well understood. This study aimed to develop reliable risk signatures associated with NK cell-related genes for predicting thyroid cancer (THCA). The single-cell RNA sequencing (scRNA-seq) data from seven THCA samples (GSE184362) and bulk-RNA-seq data of 502 THCA patients (TCGA-THCA) were included. The scRNA-seq data was analyzed using the "Seurat" R package to identify differentially expressed genes in NK cells. The clustering analysis was carried out using the R package "ConsensusClusterPlus". The gene set variation analysis (GSVA) algorithm was applied to assess the variations in biological pathways among subtypes. The ESTIMATE algorithm was utilized to calculate the scores for stromal, immune and estimate variables. In addition, we used the single sample Gene Set Enrichment Analysis and CIBERSORT algorithms to assess the degree to which immune cells and pathways related to immunity were enriched based on the meta-cohort. In the TCGA-THCA cohort, the "glmnet" R package was used for the gene selection, and LASSO Cox analysis was used to construct prognostic features. The "maftools" R package was used to examine the somatic mutation landscape of THCA in both low- and high-risk groups. One-hundred and eighty-five NK cell marker genes were screened, and nine genes were associated with the THCA prognosis. KLF2, OSTF1 and TAPBP were finally identified and constructed a risk signature with significant prognostic value. KLF2 and OSTF1 were protective genes, and TAPBP was a risk gene. Patients at high risk had a considerably lower overall survival compared with those at low risk. Mutations in the TCGA-THCA cohort were predominantly C > T. Increased tumor mutation burden (TMB) levels were linked to overall survival. The low-risk H-TMB+ group had a better prognosis, while the high-risk L-TMB+ group had the worst prognosis. Natural killer cell-related genes KLF2, OSTF1 and TAPBP were used to develop a novel prognostic risk signature, offering a new perspective on the prognosis and treatment of THCA.

Single Solution Sphere-Decoding Algorithm Model Predictive Control for High-Current Applications

Single Solution Sphere-Decoding Algorithm Model Predictive Control for High-Current Applications

Hyperparameters optimization for the machine learning

There is no single best Hyperparameter Optimization algorithm. Different optimization algorithms meet different Hyperparameter Optimization tasks with different constraints. To accelerate the Hyperparameter Optimization, it is necessary to parallelize training executions of various tests, introduce distributed training and prematurely stop unpromising tests. It is recommended to use a library approach to support the hyperparameter optimization service. Ray Tune is the best by far among the open source Hyperparameter Optimization. Hyperparameter optimisation (HPO) is the process of identifying a set of hyperparameters that yields an optimal model. By optimal, we mean the model that minimises a predefined loss function on a given data set. This is a repeated model training process, except that the neural network is trained with a different set of hyperparameters each time. In this process, the optimal set of hyperparameters will be identified. During the grid search, users specify a limited set of values for each hyperparameter and then select trial hyperparameters from the Cartesian product of these values. Once the grid is constructed, the GPGs are tested with the grid values. The grid search suffers when the number of hyperparameters becomes larger or the parameter search space becomes larger, as in this case the required number of estimates will grow exponentially. Another problem with grid search is its inefficiency. Since grid search treats each set of hyperparameter candidates equally, it will spend a lot of computational resources in the suboptimal configuration space without spending enough computational power on the optimal space.

Inversion model of soil salinity in alfalfa covered farmland based on sensitive variable selection and machine learning algorithms.

Timely and accurate monitoring of soil salinity content (SSC) is essential for precise irrigation management of large-scale farmland. Uncrewed aerial vehicle (UAV) low-altitude remote sensing with high spatial and temporal resolution provides a scientific and effective technical means for SSC monitoring. Many existing soil salinity inversion models have only been tested by a single variable selection method or machine learning algorithm, and the influence of variable selection method combined with machine learning algorithm on the accuracy of soil salinity inversion remain further studied. Firstly, based on UAV multispectral remote sensing data, by extracting the spectral reflectance of each sampling point to construct 30 spectral indexes, and using the pearson correlation coefficient (PCC), gray relational analysis (GRA), variable projection importance (VIP), and support vector machine-recursive feature elimination (SVM-RFE) to screen spectral index and realize the selection of sensitive variables. Subsequently, screened and unscreened variables as model input independent variables, constructed 20 soil salinity inversion models based on the support vector machine regression (SVM), back propagation neural network (BPNN), extreme learning machine (ELM), and random forest (RF) machine learning algorithms, the aim is to explore the feasibility of different variable selection methods combined with machine learning algorithms in SSC inversion of crop-covered farmland. To evaluate the performance of the soil salinity inversion model, the determination coefficient (R2), root mean square error (RMSE) and performance deviation ratio (RPD) were used to evaluate the model performance, and determined the best variable selection method and soil salinity inversion model by taking alfalfa covered farmland in arid oasis irrigation areas of China as the research object. The variable selection combined with machine learning algorithm can significantly improve the accuracy of remote sensing inversion of soil salinity. The performance of the models has been improved markedly using the four variable selection methods, and the applicability varied among the four methods, the GRA variable selection method is suitable for SVM, BPNN, and ELM modeling, while the PCC method is suitable for RF modeling. The GRA-SVM is the best soil salinity inversion model in alfalfa cover farmland, with Rv 2 of 0.8888, RMSEv of 0.1780, and RPD of 1.8115 based on the model verification dataset, and the spatial distribution map of soil salinity can truly reflect the degree of soil salinization in the study area. Based on our findings, the variable selection combined with machine learning algorithm is an effective method to improve the accuracy of soil salinity remote sensing inversion, which provides a new approach for timely and accurate acquisition of crops covered farmland soil salinity information.

Identification of Hypoxia and Mitochondrial-related Gene Signature and Prediction of Prognostic Model in Lung Adenocarcinoma.

Background: The correlation between hypoxia and tumor development is widely acknowledged. Meanwhile, the foremost organelle affected by hypoxia is mitochondria. This study aims to determine whether they possess prognostic characteristics in lung adenocarcinoma (LUAD). For this purpose, a bioinformatics analysis was conducted to assess hypoxia and mitochondrial scores related genes, resulting in the successful establishment of a prognostic model. Methods: Using the single sample Gene Set Enrichment Analysis algorithm, the hypoxia and mitochondrial scores were computed. Differential expression analysis and weighted correlation network analysis were employed to identify genes associated with hypoxia and mitochondrial scores. Prognosis-related genes were obtained through univariate Cox regression, followed by the establishment of a prognostic model using least absolute shrinkage and selection operator Cox regression. Two independent validation datasets were utilized to verify the accuracy of the prognostic model using receiver operating characteristic and calibration curves. Additionally, a nomogram was employed to illustrate the clinical significance of this study. Results: 318 differentially expressed genes associated with hypoxia and mitochondrial scores were identified for the construction of a prognostic model. The prognostic model based on 16 genes, including PKM, S100A16, RRAS, TUBA4A, PKP3, KCTD12, LPGAT1, ITPRID2, MZT2A, LIFR, PTPRM, LATS2, PDIK1L, GORAB, PCDH7, and CPED1, demonstrates good predictive accuracy for LUAD prognosis. Furthermore, tumor microenvironments analysis and drug sensitivity analysis indicate an association between risk scores and certain immune cells, and a higher risk scores suggesting improved chemotherapy efficacy. Conclusion: The research established a prognostic model consisting of 16 genes, and a nomogram was developed to accurately predict the prognosis of LUAD patients. These findings may contribute to guiding clinical decision-making and treatment selection for patients with LUAD, ultimately leading to improved treatment outcomes.

Community Detection in Censored Hypergraph

Community detection refers to the problem of clustering the nodes of a network (either graph or hypergrah) into groups. Various algorithms are available for community detection and all these methods apply to uncensored networks. In practice, a network may has censored (or missing) values and it is shown that censored values have non-negligible effect on the structural properties of a network. In this paper, we study community detection in censored $m$-uniform hypergraph from information-theoretic point of view. We derive the information-theoretic threshold for exact recovery of the community structure. Besides, we propose a polynomial-time algorithm to exactly recover the community structure up to the threshold. The proposed algorithm consists of a spectral algorithm plus a refinement step. It is also interesting to study whether a single spectral algorithm without refinement achieves the threshold. To this end, we also explore the semi-definite relaxation algorithm and analyze its performance.

An integrated fuse information fusion method for infrared imaging detector and range finder based on improved particle filter

AbstractAiming at the traditional integrated fuse information fusion method that cannot simultaneously obtain high‐precision burst control parameters such as burst delay time, azimuth angle, and detection angle, the infrared imaging detector and range finder information fusion method based on particle filtering with BAS algorithm is proposed. Under the centralized information fusion structure that can obtain higher accuracy, first, use a faster linear interpolation method to realize the time alignment of the two asynchronous sensors. Second, for the strong non‐linear observation model in the missile target rendezvous model, use a particle filter framework that is not affected by the nonlinearity of the system function and non‐gaussian noise. Finally, Utilizing the single search BAS algorithm with low computational complexity to solve the problem of particle degradation in particle filtering and improve the accuracy of burst control. The experiment shows that in a typical simulation environment, the optimized information fusion method improves the burst delay time accuracy by 61.82%, azimuth accuracy by 66.28%, detection angle accuracy by 36.15%, and simulation time is shorter compared to traditional methods.

An Ensemble Machine Learning Model to Detect Tax Fraud: Conceptual Framework

Most governments throughout the world, especially in developing and underdeveloped countries, depend more on tax revenue to fund public expenditure and investments. In the wake of Covid19, even governments that did not depend largely on tax revenue are forced to do that since their other sources of income were affected by the pandemic as borders were closed and nations were on lockdowns. Research, however, has shown that tax fraud is rampant especially in less developed countries. Traditional methods of detecting tax fraud are costly and they largely depend on the experts’ past experience. This renders them less effective where new mechanisms of tax fraud are involved. In this work I provide a conceptual framework on the use of ensemble machine learning models to detect tax fraud. I use decision trees, support vector machines and logistic regression as the base models. I hypothesize that ensemble methods outperform unsupervised machine learning models and the use of a single algorithm under supervised machine learning models. The outcomes of this research will serve to provide a framework that will help tax authorities to detect tax fraud thereby increasing the revenue collected.

Link Prediction through Ensemble Techniques

Link prediction is significant in the field of network analysis which helps to determine the future or hidden connections between the entities in a network. Traditional link prediction methods typically utilize single algorithms to make predictions, which might not be optimal for diverse and complex networks. In this research paper, a new framework is proposed which utilizes the existing link prediction algorithms with ensemble techniques. We have explored three ensemble strategies: bagging, boosting, and voting classifier and four individual link prediction algorithms: Adamic Adar, Katz, CCPA, Accurate link prediction. Methods considering ensemble techniques combine multiple individual models to improve prediction accuracy and robustness. We have experimentally compared the performance of proposed framework using six real world datasets in terms of AUC and Accuracy. The results show the superiority of the proposed framework over individual link prediction algorithms.

Поздние преждевременные спонтанные роды: версии и контраверсии

Aim. To study modern ideas about late premature spontaneous labor, identify unresolved issues and identify tasks for further research on this theme. Key points. The problem of late premature labor occupies a leading position in modern obstetrics and perinatology. Currently, there is no single opinion and no identical approaches in predicting late premature spontaneous labor. This prevents the creation of a single algorithm of actions to prevent them. The article presents up-to-date information on the marker serum proteins and transcriptomic analysis of the placenta, as well as highlights some aspects of the pathogenesis of premature labor on the model of fetoplacental homeostasis disorders. Conclusion. To reduce the risk of adverse outcomes of late premature spontaneous labor, it is necessary to create clear criteria for their prediction and early diagnosis, as well as timely implementation of organizational and therapeutic measures for this obstetric complication. Keywords: late premature spontaneous labor, marker serum proteins, pregnancy-associated plasma protein-A, placental growth factor, soluble fms-like tyrosine kinase-1, gene expression.

Comparison and Efficiency of Different Algorithms to Count and Determine the Composition of All Possible Riffle Shuffles Used in Magic

As part of a Technical and Scientific Project offered to L2 students of ISEP over a period of four months at the rate of one hour per week supervised, a study of the composition and the total number of all riffle shuffles used in magic was requested. The riffle shuffle consists of taking a deck of cards, cutting it into any two parts, then inserting randomly the cards of the first packet into the second. In first approximation there are 52! possible riffle shuffles. This number is an upper limit since the cards in a subdeck cannot change their relative position to their neighbours. A card A below a card B in the same subdeck will not be able to find itself above in the final shuffle. The first goal of the project is to compare one single algorithm implemented either recursively or iteratively in terms of execution time depending on the number of cards. The second goal is to compare the implementation of the same algorithm (either iterative or recursive) in several different programming languages (Python, Language C, Matlab, etc.) and to compare execution times according to the used language. Thus, a recursive algorithm was proposed to the students. All the students tried to code this algo and make it iterative … except one! He imagined another algorithm. The algorithm of the student and the teacher were compared in terms of execution time: do the student exceed the master? The aim of this article is to share my experience. I leave it to you to imagine educational sequences around this project in adequacy with your desires and the level of your students.

Comparing the Performance of Machine Learning and Deep Learning Algorithms in Wastewater Treatment Process

This study assessed the performance of single and modified algorithms based on machine learning and deep learning for wastewater treatment process. More specifically, this study adopted support vector machine (SVM), random forest (RF), and artificial neural network (ANN) for machine learning as well as long short-term memory (LSTM) for deep learning. The performance of these (single) algorithms were compared with that of modified ones processed through hyperparameter tuning, ensemble learning (only for machine learning), and multi-layer stacking (i.e., two layers of LSTM units). The daily effluent of wastewater treatment process observed between 2017 and 2022 in the Cheong-Ju National Industrial Complex was used as input to all tested algorithms, which was evaluated with respect to mean squared error. For the model performance evaluation, discharge and biochemical oxygen demand are selected as dependent variables out of nine measured parameters. Results showed that the performance of any machine learning algorithms was superior to their competitor LSTM. This is mainly attributed to a small amount of input data provided to the LSTM algorithm and unstable effluent wastewater characteristics. Meanwhile, hyperparameter tuning improved the performance of all tested algorithms. However, ensemble learning for machine learning and two-layer stacking for LSTM generally resulted in performance degradation as compared to that of single algorithms, regardless of dependent variables. Therefore, this calls for a careful design and evaluation of modified algorithms, specifically for model architecture and performance improvement processes.