Gaussian Mixed Model Research Articles

Can the digital economy improve green total factor productivity? An empirical study based on Chinese urban data.

With the new generation of technological revolution, the digital economy has progressively become a key driver of global economic development. In this context, how to promote green economic growth and improve green total factor productivity (GTFP) with the help of the digital economy is an important issue that urgently needs empirical research. We adopted the panel data of 278 Chinese prefecture-level cities from 2011 to 2020 to test whether the digital economy improves the GTFP through the Gaussian Mixed Model (GMM) dynamic panel model. The moderating effect model has been used to explore the impact mechanism from the perspectives of industrial structure upgrade and environmental regulation. In addition, a grouping regression was applied to the sample cities to test the heterogeneous impact of the digital economy on the GTFP. Based upon the empirical findings, this work has the following conclusions. First, the digital economy plays a significant role in improving the GTFP. Second, an industrial structure upgrade has a positive moderating effect on the ability of the digital economy to enhance the GTFP. The environmental regulation, in contrast, has a negative moderating effect. Third, the digital economy exerts heterogeneous impacts on the GTFP across regions, but not at the city level.

A new Monte Carlo sampling method based on Gaussian Mixture Model for imbalanced data classification.

Imbalanced data classification has been a major topic in the machine learning community. Different approaches can be taken to solve the issue in recent years, and researchers have given a lot of attention to data level techniques and algorithm level. However, existing methods often generate samples in specific regions without considering the complexity of imbalanced distributions. This can lead to learning models overemphasizing certain difficult factors in the minority data. In this paper, a Monte Carlo sampling algorithm based on Gaussian Mixture Model (MCS-GMM) is proposed. In MCS-GMM, we utilize the Gaussian mixed model to fit the distribution of the imbalanced data and apply the Monte Carlo algorithm to generate new data. Then, in order to reduce the impact of data overlap, the three sigma rule is used to divide data into four types, and the weight of each minority class instance based on its neighbor and probability density function. Based on experiments conducted on Knowledge Extraction based on Evolutionary Learning datasets, our method has been proven to be effective and outperforms existing approaches such as Synthetic Minority Over-sampling TEchnique.

A fine extraction algorithm for image-based surface cracks in underwater dams

Under the action of water erosion and self-aging, reservoir dams are prone to develop cracks, which affect safe operation. Underwater visual imaging can be used to detect dam surface cracks, but spalling, aquatic plants and suspended sediments result in low image contrast and complex backgrounds. With the use of unsupervised machine learning, this paper proposes a fine segmentation and extraction algorithm for image-based dam surface cracks. First, adaptive histogram equalization is used to change the uneven illumination areas of underwater surface images intoeven illumination areas, whose statistical characteristics are calculated under linear spatial filtering. Second, the extraction problem of crack areas of interest after dodging preprocessing is transformed into calculating the distance of the image block cluster center, which can distinguish the image blocks of crack features from the background interference features. Third, the fine extraction of crack images is carried out by considering the connected domains and morphological features, and the posterior probability of an image sample category is obtained based on the soft clustering of a Gaussian mixed model. Finally, different extraction algorithms related to surface cracks are evaluated in extensive experiments. The results validate the superior performance of the proposed extraction algorithm with 90.1% extraction accuracy, 6.5% missing alarm rate and 7.2% false alarm rate.

Bayesian Inference for Mixed Gaussian GARCH-Type Model by Hamiltonian Monte Carlo Algorithm.

MCMC algorithm is widely used in parameters' estimation of GARCH-type models. However, the existing algorithms are either not easy to implement or not fast to run. In this paper, Hamiltonian Monte Carlo (HMC) algorithm, which is easy to perform and also efficient to draw samples from posterior distributions, is firstly proposed to estimate for the Gaussian mixed GARCH-type models. And then, based on the estimation of HMC algorithm, the forecasting of volatility prediction is investigated. Through the simulation experiments, the HMC algorithm is more efficient and flexible than the Griddy-Gibbs sampler, and the credibility interval of forecasting for volatility prediction is also more accurate. A real application is given to support the usefulness of the proposed HMC algorithm well.

Resource Sharing Method of Basic Computer Education Based on Mixed Gaussian Model

This study developed a computer-based basic educational resource sharing method based on a mixed Gaussian model in order to improve the effect of educational resource sharing. Initially, the challenges of educational resource sharing are examined. Then, the basic sharing platform, consisting of the index layer, the computer basic education resource library, and the data layer, is designed. After collecting information about basic computer education resources to be shared, the education resources are clustered using the Gaussian mixture model and then configured based on the results of the clustering. Depending on the number of configuration factors, computer basic education resources can be shared in accordance with their design. The experimental results indicate that this method has high resource sharing efficiency, the ability to update resources in real time, and a good teacher and student evaluation system. Compared to traditional sharing methods, the proposed method has the benefits of a shorter sharing time and greater resource clustering precision.

PSVIII-B-15 Potential Involvement of DNA Methylation of First Intron in Transcriptional Regulation During Bovine Subclinical Mastitis Caused by Staphylococcus Aureus

Abstract DNA methylation involvement in the regulation of mammary gland inflammatory defense against pathogens, including Staphylococcus aureus(SA) has been documented. However, the mechanisms are not clear. The possible influence of DNA methylation of first introns (DNA-MeFI) on transcriptional activities during subclinical mastitis was studied by profiling the methylome (whole genome methylation sequencing) and transcriptome (RNA-sequencing) of milk somatic cells from cows(n=15) with SA subclinical mastitis (SACs) and healthy cows (HCs)(n=13), followed by bioinformatics processing with standard tools. The DNA-MeFI was calculated, for each of 19411 genes found as expressed in the samples, by averaging the methylation levels of all CpG sites in the first intron. At genome-wide, the DNA-MeFIs was inversely correlated with gene expression (GE) (Pearson’s r=-0.173, p=1.60×10-122)(Figure 1-A). Similarly, the difference in the DNA-MeFI and GE between SACs and HCs was also inversely correlated (Pearson’s r=-0.149, p=4.25×10-97)(Figure 1-B). Application of machine learning by Gaussian Mixed model (GMM) using scikit-learn in python to the changes of DNA-MeFI and GE between SACs and HCs revealed 2866 outliers (GMM’s p< 0.005) that showed significant changes in the DNA-MeFI and/or GE. 644 genes with >10% difference in DNA-MeFI and |log2FC| >1 in GE were selected and referred to as differentially methylated and expressed genes (DME-genes). 644 DME-genes were significantly enriched in 5 GO terms and 11 KEGG pathways related to diseases and immune functions, including Staphylococcus aureus infection and Natural killer cell mediated cytotoxicity (Table 1). Furthermore, 74.84% DME-genes showed inverse changes in DNA-MeFI and GE. Besides, the DNA-MeFI of 264 DME-genes were found to correlate significantly with their GE (|rho| >0.3, FDR<0.05), including 225 DME-genes(85.23%) with inverse correlations(rho< -0.3, FDR< 0.05). 39 DME-genes with positive correlations were significantly enriched in 21 GO terms mainly related to metabolic processes. In conclusion, the DNA-MeFI possibly participate in the regulation of gene expression during bovine subclinical mastitis caused by SA.

Crack classification of fiber-reinforced backfill based on Gaussian mixed moving average filtering method

The crack classification in the rupture process of filling materials is important for predicting rupture instability and identifying rupture precursor features. To investigate fiber-reinforced backfill crack classification, uniaxial compression tests were performed on different fiber-reinforced backfills using a microcomputer-controlled electronic universal testing machine and a Micro-II Express Digital acoustic emission (AE) system. The crack classification model based on the Gaussian mixed moving average filtering method was combined with AE technique under uniaxial compression test, and the crack classification results of the Japanese Construction Materials Standard (JCMS-III B5706) parametric analysis method and the Gaussian mixed model (GMM) were compared and analyzed by comparing changes in the distribution of RA (rise time/maximum amplitude) and AF (average frequency) values. The results revealed that the fiber-reinforced backfill is dominated by tensile cracks in the initial loading stage, tensile cracks transitioned to shear cracks in the intermediate loading stage, and shear cracks dominated in the final loading stage. The fiber-reinforced backfill exhibits an overall distribution of high RA values and low AF values, and the data points are generally close to the RA half-axis, increasing along the RA half-axis in a striped distribution. The percentages of tensile and shear cracks obtained based on the GMM are roughly the same as those obtained by the JCMS-III B5706 parametric analysis method, indicating that the model is feasible to be applied to the crack classification of fiber-reinforced backfills. The results of the study can provide a new method for the crack classification during the rupture of fiber-reinforced filling materials in mines.

Causal discoveries for high dimensional mixed data.

Causal relationships are of crucial importance for biological and medical research. Algorithms have been proposed for causal structure learning with graphical visualizations. While much of the literature focuses on biological studies where data often follow the same distribution, for example, the normal distribution for all variables, challenges emerge from epidemiological and clinical studies where data are often mixed with continuous, binary, and ordinal variables. We propose to use a mixed latent Gaussian copula model to estimate the underlying correlation structure via the rank correlation for mixed data. This correlation structure is then incorporated into a popular causal discovery algorithm, the PC algorithm, to identify causal structures. The proposed algorithm, called the latent-PC algorithm, is able to discover the true causal structure consistently under mild conditions in high dimensional settings. From simulation studies, the latent-PC algorithm delivers a competitive performance in terms of a similar or higher true positive rate and a similar or lower false positive rate, compared with other variants of the PC algorithm. In the high dimensional settings where the number of variables is more than the number of observations, the causal graphs identified by the latent-PC algorithm are closer to the true causal structures, compared to other competing algorithms. Further, we demonstrate the utility of the latent-PC algorithm in a real dataset for hepatocellular carcinoma. Causal structures for patient survival are visualized and connected with clinical interpretations in the literature.

Dual-Anchor Metric Learning for Blind Image Quality Assessment of Screen Content Images

The natural scene statistic is destroyed by the artificial portion in the screen content images (SCIs) and is also impractical for obtaining an accurate statistical model due to the variable composition of the artificial and natural parts in SCIs. To resolve this problem, this paper presents a dual-anchor metric learning (DAML) method that is inspired by metric learning to obtain discriminative statistical features and further identify complex distortions, as well as predict SCI image quality. First, two Gaussian mixed models with prior data are constructed as the target anchors of the statistical model from natural and artificial image databases, which can effectively enhance the metrical discrimination of the mapping relation between the feature representation and quality degradation by conditional probability analysis. Then, the distances of the high-order statistics are softly aggregated to conduct metric learning between the local features and clusters of each target statistical model. Through empirical analysis and experimental verification, only variance differences are used as quality-aware features to benefit the balance of complexity and effectiveness. Finally, the mapping model between the target distances and subjective quality can be obtained by support vector regression. To validate the performance of DAML, multiple experiments are carried out on three public databases: SIQAD, SCD, and SCID. Meanwhile, PLCC, SRCC, and the RMSE are then employed to compute the correlation between subjective and objective ratings, which can estimate the prediction of accuracy, monotonicity, and consistency, respectively. The PLCC and RMSE of the method achieved 0.9136 and 0.7993. The results confirm the good performance of the proposed method.

Identification of Candidate Genes Associated With Prognosis in Glioblastoma.

BackgroundGlioblastoma (GBM) is the most common malignant primary brain tumor, which associated with extremely poor prognosis.MethodsData from datasets GSE16011, GSE7696, GSE50161, GSE90598 and The Cancer Genome Atlas (TCGA) were analyzed to identify differentially expressed genes (DEGs) between patients and controls. DEGs common to all five datasets were analyzed for functional enrichment and for association with overall survival using Cox regression. Candidate genes were further screened using least absolute shrinkage and selection operator (LASSO) and random forest algorithms, and the effects of candidate genes on prognosis were explored using a Gaussian mixed model, a risk model, and concordance cluster analysis. We also characterized the GBM landscape of immune cell infiltration, methylation, and somatic mutations.ResultsWe identified 3,139 common DEGs, which were associated mainly with PI3K-Akt signaling, focal adhesion, and Hippo signaling. Cox regression identified 106 common DEGs that were significantly associated with overall survival. LASSO and random forest algorithms identified six candidate genes (AEBP1, ANXA2R, MAP1LC3A, TMEM60, PRRG3 and RPS4X) that predicted overall survival and GBM recurrence. AEBP1 showed the best prognostic performance. We found that GBM tissues were heavily infiltrated by T helper cells and macrophages, which correlated with higher AEBP1 expression. Stratifying patients based on the six candidate genes led to two groups with significantly different overall survival. Somatic mutations in AEBP1 and modified methylation of MAP1LC3A were associated with GBM.ConclusionWe have identified candidate genes, particularly AEBP1, strongly associated with GBM prognosis, which may help in efforts to understand and treat the disease.

Impact of Fiscal Expenditure on Farmers’ Livelihood Capital in the Ethnic Minority Mountainous Region of Sichuan, China

Poverty alleviation is the first battle to realize the rural revitalization strategy in China, and research on the sustainable livelihood of rural households is of great significance to solving the problem of rural poverty. Based on the sustainable livelihood framework, the reasonable scale and structure of fiscal expenditure is an inevitable requirement in producing sustainable livelihood capital toward this aim. In this study, the system Gaussian mixed model was used to analyze the impact of fiscal expenditure on farmers’ livelihoods. Representative survey data uses panel data from 48 counties across Liangshan Prefecture, Ganzi Prefecture, and Aba Prefecture in Sichuan, China. The results are as follows: (1) The average stock of human capital in 2010 to 2015 was the highest in the composition of farmers’ livelihood capital; (2) natural capital and physical capital were positively affected by the total scale of fiscal expenditure, agriculture, forestry, and water expenditure, and the former was negatively affected by general public service expenditure, education expenditure, social security and employment expenditure, and medical expenditure; (3) financial capital and the total amount of livelihood capital were positively affected by the total scale of fiscal expenditure, agriculture, forestry and water expenditure, education expenditure, social security and employment expenditure, and medical expenditure, and negatively affected by general public service expenditure; (4) human capital was positively affected by the total scale of fiscal expenditure, education expenditure, social security and employment expenditure, and medical expenditure; and (5) social capital was positively affected by agriculture, forestry and water expenditure, and education expenditure.

A model-based clustering of expectation–maximization and K -means algorithms in crime hotspot analysis

Hotspot analysis of spatial attributes is a persistent research field in data mining, and applying a model-based clustering procedure is increasingly becoming popular in identifying trends and patterns in datasets on crime events occurring in space. The distributions of potential crime hotspots are parameterized as arising from Gaussian multivariate distributions, whose parameters are estimated by the expectation–maximization (E-M) algorithm, an iterative process with convergence very sensitive to initializations. In this study, a model-based clustering algorithm is explored from the E-M algorithm, initialized by K-means clustering using geodesic distance classification to estimate the model parameters and compared with the classical E-M algorithm, initialized with hierarchical clustering, to identify the distributional patterns of incidence of criminal activities. These model-based clustering algorithms were demonstrated on an open-source large dataset of violent crime activities, which occurred in West Midlands County. Training the data as a Gaussian process, the study identified 12 hotspots of Gaussian mixed models as clusters of an ellipsoidal distribution varying in shape, volume, and orientation, which are mostly found in central parts of boroughs of the study area. The proposed model-based clustering of the E-M algorithm combined with K-means clustering algorithm proved efficient as being fast and stable in convergence with low probability of uncertainty by classifications, producing same classification in some cases when compared to that of the classical E-M and K-means algorithms. The combined model-based clustering techniques applied in the hotspot analysis of criminal activities in space will not only provide insight into crime prediction and resource allocation in combating strategies but also guide researchers to adopt mechanisms for modeling large spatial attributes in data mining.

Wrist pulse signal based vascular age calculation using mixed Gaussian model and support vector regression.

Vascular age (VA) is the direct index to reflect vascular aging, so it plays a particular role in public health. How to obtain VA conveniently and cheaply has always been a research hotspot. This study proposes a new method to evaluate VA with wrist pulse signal. Firstly, we fit the pulse signal by mixed Gaussian model (MGM) to extract the shape features, and adopt principal component analysis (PCA) to optimize the dimension of the shape features. Secondly, the principal components and chronological age (CA) are respectively taken as the independent variables and dependent variable to establish support vector regression (SVR) model. Thirdly, the principal components are fed into the SVR model to predicted the vascular aging of each subject. The predicted value is regarded as the description of VA. Finally, we compare the correlation coefficients of VA with pulse width (PW), inflection point area ratio (IPA), Ratio b/a (RBA), augmentation index (AIx), diastolic augmentationindex (DAI) and pulse transit time (PTT) with those of CA with these six indices. Compared with the CA, the VA is closer to PW (r = 0.539, P < 0.001 to r = 0.589, P < 0.001 in men; r = 0.325, P < 0.001 to r = 0.400, P < 0.001 in women), IPA (r = - 0.446, P < 0.001 to r = - 0.534, P < 0.001 in men; r = - 0.623, P < 0.001 to r = - 0.660, P < 0.001 in women), RBA (r = 0.328, P < 0.001 to r = 0.371, P < 0.001 in women), AIx (r = 0.659, P < 0.001 to r = 0.738, P < 0.001 in men; r = 0.547, P < 0.001 to r = 0.573, P < 0.001 in women), DAI (r = 0.517, P < 0.001 to r = 0.532, P < 0.001 in men; r = 0.507, P < 0.001 to r = 0.570, P < 0.001 in women) and PTT (r = 0.526, P < 0.001 to r = 0.659, P < 0.001 in men; r = 0.577, P < 0.001 to r = 0.814, P < 0.001 in women). The VA is more representative of vascular aging than CA. The method presented in this study provides a new way to directly and objectively assess vascular aging in public health.

Decomposition of variation of mixed variables by a latent mixed Gaussian copula model.

Many biomedical studies collect data of mixed types of variables from multiple groups of subjects. Some of these studies aim to find the group-specific and the common variation among all these variables. Even though similar problems have been studied by some previous works, their methods mainly rely on the Pearson correlation, which cannot handle mixed data. To address this issue, we propose a latent mixed Gaussian copula (LMGC) model that can quantify the correlations among binary, ordinal, continuous, and truncated variables in a unified framework. We also provide a tool to decompose the variation into the group-specific and the common variation over multiple groups via solving a regularized M-estimation problem. We conduct extensive simulation studies to show the advantage of our proposed method over the Pearson correlation-based methods. We also demonstrate that by jointly solving the M-estimation problem over multiple groups, our method is better than decomposing the variation group by group. We also apply our method to a Chlamydia trachomatis genital tract infection study to demonstrate how it can be used to discover informative biomarkers that differentiatepatients.

Vehicle Detection for Vision-Based Intelligent Transportation Systems Using Convolutional Neural Network Algorithm

Vehicle detection in Intelligent Transportation Systems (ITS) is a key factor ensuring road safety, as it is necessary for the monitoring of vehicle flow, illegal vehicle type detection, incident detection, and vehicle speed estimation. Despite the growing popularity in research, it remains a challenging problem that must be solved. Hardware-based solutions such as radars and LIDAR are been proposed but are too expensive to be maintained and produce little valuable information to human operators at traffic monitoring systems. Software based solutions using traditional algorithms such as Histogram of Gradients (HOG) and Gaussian Mixed Model (GMM) are computationally slow and not suitable for real-time traffic detection. Therefore, the paper will review and evaluate different vehicle detection methods. In addition, a method of utilizing Convolutional Neural Network (CNN) is used for the detection of vehicles from roadway camera outputs to apply video processing techniques and extract the desired information. Specifically, the paper utilized the YOLOv5s architecture coupled with k-means algorithm to perform anchor box optimization under different illumination levels. Results from the simulated and evaluated algorithm showed that the proposed model was able to achieve a mAP of 97.8 in the daytime dataset and 95.1 in the nighttime dataset.

Imaging and characterization of transitions in biofilm morphology via anomalous diffusion following environmental perturbation.

Microorganisms form macroscopic structures for the purpose of environmental adaptation. Sudden environmental perturbations induce dynamics that cause bacterial biofilm morphology to transit to another equilibrium state, thought to be related to anomalous diffusion processes. Here, detecting the super-diffusion characteristics would offer a long-sought goal for a rapid detection method of biofilm phenotypes based on their dynamics, such as growth or dispersal. In this paper, phase-sensitive Doppler optical coherence tomography (OCT) and dynamic light scattering (DLS) are combined to demonstrate wide field-of-view and label-free internal dynamic imaging of biofilms. The probability density functions (PDFs) of phase displacement of the backscattered light and the dynamic characteristics of the PDFs are estimated by a simplified mixed Cauchy and Gaussian model. This model can quantify the super-diffusion state and estimate the dynamic characteristics and macroscopic responses in biofilms that may further describe dispersion and growth in biofilm models.

A Novel Trajectory Based Prediction Method for Urban Subway Design

In recent years, with the development of various types of public transportation, they are also more and more closely connected. Among them, subway transportation has become the first choice of major cities. However, the planning of subway stations is very difficult and there are many factors to consider. Besides, few methods for selecting optimal station locations take other public transport in to consideration. In order to study the relationship between different types of public transportation, the authors collected and analyzed the travel data of subway passengers and the passenger trajectory data of taxis. In this paper, a method based on LeaderRank and Gaussian Mixed Model (GMM) is proposed to conduct subway station locations selection. In this method, the author builds a subway-passenger traffic zone weighted network and a station location prediction model. First, we evaluate the nodes in the network, then use the GPS track data of taxis to predict the location of new stations in future subway construction, and analyze and discuss the land use characteristics in the prediction area. Taking the design of the Beijing subway line as an example, the suitability of this method is illustrated.

The importance of association of comorbidities on COVID-19 outcomes: a machine learning approach

Background The individual influence of a variety of comorbidities on COVID-19 patient outcomes has already been analyzed in previous works in an isolated way. We aim to determine if different associations of diseases influence the outcomes of inpatients with COVID-19. Methods Retrospective cohort multicenter study based on clinical practice. Data were taken from the SEMI-COVID-19 Registry, which includes most consecutive patients with confirmed COVID-19 hospitalized and discharged in Spain. Two machine learning algorithms were applied in order to classify comorbidities and patients (Random Forest -RF algorithm, and Gaussian mixed model by clustering -GMM-). The primary endpoint was a composite of either, all-cause death or intensive care unit admission during the period of hospitalization. The sample was randomly divided into training and test sets to determine the most important comorbidities related to the primary endpoint, grow several clusters with these comorbidities based on discriminant analysis and GMM, and compare these clusters. Results A total of 16,455 inpatients (57.4% women and 42.6% men) were analyzed. According to the RF algorithm, the most important comorbidities were heart failure/atrial fibrillation (HF/AF), vascular diseases, and neurodegenerative diseases. There were six clusters: three included patients who met the primary endpoint (clusters 4, 5, and 6) and three included patients who did not (clusters 1, 2, and 3). Patients with HF/AF, vascular diseases, and neurodegenerative diseases were distributed among clusters 3, 4 and 5. Patients in cluster 5 also had kidney, liver, and acid peptic diseases as well as a chronic obstructive pulmonary disease; it was the cluster with the worst prognosis. Conclusion The interplay of several comorbidities may affect the outcome and complications of inpatients with COVID-19.

Analysis of the Correlation between Football Education Environment and Students' Psychology Health Based on Gauss Characteristics.

Campus football has become a core content of school physical education. Through football education, we can cultivate students' sound personality and promote students' all-round physical and mental development. At the same time, through psychological skills training methods, we can enrich the educational methods of football skills and provide theoretical reference for promoting educational reform. On the basis of Gaussian features, this paper combines the mixed Gaussian feature model to further describe the relationship between football education and students' psychology. At the same time, Apriori association rule algorithm in data mining is introduced, and Apriori algorithm is improved in parallel with Hadoop data processing platform. Several parallel association rule algorithms are emphatically studied and analyzed to strengthen the analysis of the relationship between football education and students' psychology. The results show that the average recognition rate of the correlation between football education and students' psychology based on Gaussian features is 17.91% higher than that of ordinary results, which obviously improves the correlation recognition result and has a good descriptive ability. Therefore, it has become an important issue for today's physical education workers to analyze the correlation between football education and students' psychology in order to cultivate students' sports and mental health.

Resident user load classification method based on improved Gaussian mixture model clustering

In view of the limitation of “hard assignment” of clusters in traditional clustering methods and the difficulty of meeting the requirements of clustering efficiency and clustering accuracy simultaneously in regard to massive data sets, a load classification method based on a Gaussian mixture model combining clustering and principal component analysis is proposed. The load data are fed into a Gaussian mixture model clustering algorithm after principal component analysis and dimensionality reduction to achieve classification of large-scale load datasets. The method in this paper is used to classify loads in the Canadian AMPds2 public dataset and is compared with K-Means, Gaussian mixed model clustering and other methods. The results show that the proposed method can not only achieve load classification more effectively and finely, but also save computational cost and improve computational efficiency.