Gaussian mixture model clustering with completed likelihood minimum message length criterion

An improved Gaussian mixture model （GMM）- based clustering method is proposed for the difficult case where the true distribution of data is against the assumed GMM. First, an improved model selection criterion, the completed likelihood minimum message length criterion, is derived. It can measure both the goodness-of-fit of the candidate GMM to the data and the goodness-of-partition of the data. Secondly, by utilizing the proposed criterion as the clustering objective function, an improved expectation- maximization （EM） algorithm is developed, which can avoid poor local optimal solutions compared to the standard EM algorithm for estimating the model parameters. The experimental results demonstrate that the proposed method can rectify the over-fitting tendency of representative GMM-based clustering approaches and can robustly provide more accurate clustering results.

Nonlinear mixed-effects height to crown base and crown length dynamic models using the branch mortality technique for a Korean larch( Larix olgensis ) plantations in northeast China

Korean larch(Larix olgensis)is one of the main tree species for aff orestation and timber production in northeast China.However,its timber quality and growth ability are largely infl uenced by crown size,structure and shape.The majority of crown models are static models based on tree size and stand characteristics from temporary sample plots,but crown dynamic models has seldom been constructed.Therefore,this study aimed to develop height to crown base(HCB)and crown length(CL)dynamic models using the branch mortality technique for a Korean larch plantation.The nonlinear mixed-eff ects model with random eff ects,variance functions and correlation structures,was used to build HCB and CL dynamic models.The data were obtained from 95 sample trees of 19 plots in Meng JiaGang forest farm in Northeast China.The results showed that HCB progressively increases as tree age,tree height growth(HT growth)and diameter at breast height growth(DBH growth).The CL was increased with tree age in 20 years ago,and subsequently stabilized.HT growth,DBH growth stand basal area(BAS)and crown competition factor(CCF)signifi cantly infl uenced HCB and CL.The HCB was positively correlated with BAS,HT growth and DBH growth,but negatively correlated with CCF.The CL was positively correlated with BAS and CCF,but negatively correlated with DBH growth.Model fi tting and validation confi rmed that the mixed-eff ects model considering the stand and tree level random eff ects was accurate and reliable for predicting the HCB and CL dynamics.However,the models involving adding variance functions and time series correlation structure could not completely remove heterogeneity and autocorrelation,and the fi tting precision of the models was reduced.Therefore,from the point of view of application,we should take care to avoid setting up over-complex models.The HCB and CL dynamic models in our study may also be incorporated into stand growth and yield model systems in China.

Effect of calibration data series length on performance and optimal parameters of hydrological model

In order to assess the effects of calibration data series length on the performance and optimal parameter values of a hydrological model in ungauged or data-limited catchments （data are non-continuous and fragmental in some catchments）, we used non-continuous calibration periods for more independent streamflow data for SIMHYD （simple hydrology） model calibration. Nash-Sutcliffe efficiency and percentage water balance error were used as performance measures. The particle swarm optimization （PSO） method was used to calibrate the rainfall-runoff models. Different lengths of data series ranging from one year to ten years, randomly sampled, were used to study the impact of calibration data series length. Fifty-five relatively unimpaired catchments located all over Australia with daily precipitation, potential evapotranspiration, and streamflow data were tested to obtain more general conclusions. The results show that longer calibration data series do not necessarily result in better model performance. In general, eight years of data are sufficient to obtain steady estimates of model performance and parameters for the SIMHYD model. It is also shown that most humid catchments require fewer calibration data to obtain a good performance and stable parameter values. The model performs better in humid and semi-humid catchments than in arid catchments. Our results may have useful and interesting implications for the efficiency of using limited observation data for hydrological model calibration in different climates.

On the Elastic Vibration Model for High Length-Diameter Ratio Rocket with Attitude Control System

An elastic vibration model for high length diameter ratio spinning rocket with attitude control system which can be used for trajectory simulation is established. The basic theory of elastic dynamics and vibration dynamics were both used to set up the elastic vibration model of rocket body. In order to study the problem more conveniently, the rocket's body was simplified to be an even beam with two free ends. The model was validated by simulation results and the test data.

Periodic Packing Mode for Trickle-Bed Reactors: Experiments and Modeling

A periodic packing mode of trickle-bed reactor (TBR) for the gas limited reaction was proposed. Hy-drogenation of 2-ethylanthraquinone over Pd/Al2O3 in a laboratory-scale TBR was taken as a test reaction for determining whether the periodic packing mode is advantageous. The effects of operating conditions and packing type on TBR performance were experimentally examined to demonstrate the cause-effect relationships. A mathe-matic model of TBR considering axial dispersion and fractional wetting was developed to quantitatively illuminate the reason of performance enhancement.

A Combined-Power Model for the Distribution of Axial Force in Shaft Anchor along Its Length

The theoretical results of axial force distribution models differ greatly from tests because of the complication of the rock type material. A three-parameter combined-power model is proposed by curves fitting the test data recorded from the pull tests on anchoring bars used in different engineering projects. Based on the comparison of the mechanical characteristics of shaft anchors and prestressed tendons, a two-parameter combined-power function model for prestressed tendons is proposed. The bounded length derived from the model and the suggested values of the parameters are also proposed. Compared with the Gaussian model, the three-parameter combined-power model is more precise and simple in expression. Results also suggest that the bounded length calculated from the average stress method is not safe enough.

Uncertainty of Slope Length Derived from Digital Elevation Models of the Loess Plateau, China

Although many studies have investigated slope gradient uncertainty derived from Digital Elevation Models(DEMs), the research concerning slope length uncertainty is far from mature. This discrepancy affects the availability and accuracy of soil erosion as well as hydrological modeling. This study investigates the formation and distribution of existing errors and uncertainties in slope length derivation based on 5-m resolution DEMs of the Loess Plateau in the middle of China. The slope length accuracy in three different landform areas is examined to analyse algorithm effects. The experiments indicate that the accuracy of the flat test area is lower than that of the rougher areas. The value from the specific contributing area(SCA) method is greater than the cumulative slope length(CSL), and the differences between these two methods arise from the shape of the upslope area. The variation of mean slope length derived from various DEM resolutions and landforms. The slope length accuracy decreases with increasing grid size and terrain complexity at the six test sites. A regression model is built to express the relationship of mean slope length with DEM resolution less than 85 m and terrain complexity represented by gully density. The results support the understanding of the slope length accuracy, thereby aiding in the effective evaluation of the modeling effect of surface process.

Cybersecurity: A Statistical Predictive Model for the Expected Path Length

The object of this study is to propose a statistical model for predicting the Expected Path Length (expected number of steps the attacker will take, starting from the initial state to compromise the security goal—EPL) in a cyber-attack. The model we developed is based on utilizing vulnerability information along with having host centric attack graph. Utilizing the developed model, one can identify the interaction among the vulnerabilities and individual variables (risk factors) that drive the Expected Path Length. Gaining a better understanding of the relationship between vulnerabilities and their interactions can provide security administrators a better view and an understanding of their security status. In addition, we have also ranked the attributable variables and their contribution in estimating the subject length. Thus, one can utilize the ranking process to take precautions and actions to minimize Expected Path Length.

Estimation on principal component of multi-collinearity Gauss-Markov model based on minimum description length

Gauss-Markov model is frequently used in data analysis; the analysis and estimation of its parameters is always a hot issue. Based on the information theory and from the viewpoint of optimal information on description—minimum description length, this paper discusses a case: where there is multi-collinearity in the coefficient matrix, principal component estimation is used to estimate and select the original parameters, so as to reduce its multi-collinearity and improve its credibility. From the viewpoint of minimum description length, this paper discusses the approach of selecting principal components and uses this approach to solve a practical problem.

Performance Prediction of Structured Packing Column for Cryogenic Air Separation with Hybrid Model

A detailed investigation of a thermodynamic process in a structured packing distillation column is of great impor- tance in prediction of process efficiency. In order to keep the simplicity of an equilibrium stage model and the accu- racy of a non-equilibrium stage model, a hybrid model is developed to predict the structured packing column in cryogenic air separation. A general solution process for the equilibrium stage model is developed to solve the set of equations of the hybrid model, in which a separation efficiency function is introduced to obtain the resulting tri-diagonal matrix and its solution by the Thomas algorithm. As an example, the algorithm is applied to analyze an upper column of a cryogenic air separation plant with the capacity of 17000 m3·h-1. Rigorous simulations are conducted using Aspen RATEFRAC module to validate the approach. The temperature and composition distributions are in a good agreement with the two methods. The effects of inlet/outlet position and flow rate on the temperature and composition distributions in the column are analyzed. The results demonstrate that the hybrid model and the solution algorithms are effective in analvzin~ the distillation process for a a cryogenic structured packing column.

One New Method to Obtain the Correlation Length of Solvable Models

We propose a new method to obtain the correlation length of gapped XXZ spin 1/2 antiferromagnetic chains. Following the relativistic quantum field theory in space-time dimensions, we use the exact dispersion of massive spinon to calculate the correlation length for XXZ spin 1/2 chain. We conjecture that the correlation length for other 1D lattice models can be obtained in the same way. Relation between dispersion and the oscillated correlation of gapped incommensurate lattice models is also discussed.

Mathematical model of determination of die bearing length in design of aluminum profile extrusion die

Based on the finite element simulation of profile extrusion process, the effect of local extrusion ratio, die bearing area and the distance between extrusion cylindrical center and local die orfice center on mental flow velocity was investigated. The laws of deformed metal flow on profile extrusion process were obtained. The smaller the local extrusion ratio, the faster the metal flow velocity; the smaller the area of die bearing, the faster the metal flow velocity; the smaller the distance of position of local die orifice(the closer the distance of position of local die orifice from extrusion cylindrical axis), the faster the metal flow velocity. The effect of main parameters of die structure on metal flow velocity was integrated and the mathematical model of determination of die bearing length in design of aluminum profile extrusion die was proposed. The calculated results with proposed model were well compared with the experimental results. The proposed model can be applied to determine die bearing length in design of aluminum profile extrusion die.

Minimum Description Length Methods in Bayesian Model Selection: Some Applications

Computations involved in Bayesian approach to practical model selection problems are usually very difficult. Computational simplifications are sometimes possible, but are not generally applicable. There is a large literature available on a methodology based on information theory called Minimum Description Length (MDL). It is described here how many of these techniques are either directly Bayesian in nature, or are very good objective approximations to Bayesian solutions. First, connections between the Bayesian approach and MDL are theoretically explored;thereafter a few illustrations are provided to describe how MDL can give useful computational simplifications.

Comparison of Slope Length Factor Extraction in Hillslope Soil Erosion Model with Different DEM Resolutions

In this study, non-cumulative slope length(NCSL) calculation method and spatial analytical calculation(SAC) method were respectively applied to extract slope length and slope length factor from 10 sample areas, which are located in Ansai County, north Shaanxi Province. The comparison of computation precision between variable DEM resolutions showed that NCSL was superior to SAC entirely. And the results were best when the DEM resolutions were 5 and 10 m. Besides, the results of slope length factor were nearly the same under the two conditions. So DEM of 10 m resolution can be used to extract slope length.

Robust Length of Stay Prediction Model for Indoor Patients

Due to unforeseen climate change,complicated chronic diseases,and mutation of viruses’hospital administration’s top challenge is to know about the Length of stay(LOS)of different diseased patients in the hospitals.Hospital management does not exactly know when the existing patient leaves the hospital;this information could be crucial for hospital management.It could allow them to take more patients for admission.As a result,hospitals face many problems managing available resources and new patients in getting entries for their prompt treatment.Therefore,a robust model needs to be designed to help hospital administration predict patients’LOS to resolve these issues.For this purpose,a very large-sized data(more than 2.3 million patients’data)related to New-York Hospitals patients and containing information about a wide range of diseases including Bone-Marrow,Tuberculosis,Intestinal Transplant,Mental illness,Leukaemia,Spinal cord injury,Trauma,Rehabilitation,Kidney and Alcoholic Patients,HIV Patients,Malignant Breast disorder,Asthma,Respiratory distress syndrome,etc.have been analyzed to predict the LOS.We selected six Machine learning(ML)models named:Multiple linear regression(MLR),Lasso regression(LR),Ridge regression(RR),Decision tree regression(DTR),Extreme gradient boosting regression(XGBR),and Random Forest regression(RFR).The selected models’predictive performance was checked using R square andMean square error(MSE)as the performance evaluation criteria.Our results revealed the superior predictive performance of the RFRmodel,both in terms of RS score(92%)and MSE score(5),among all selected models.By Exploratory data analysis(EDA),we conclude that maximumstay was between 0 to 5 days with the meantime of each patient 5.3 days and more than 50 years old patients spent more days in the hospital.Based on the average LOS,results revealed that the patients with diagnoses related to birth complications spent more days in the hospital than other diseases.This finding could help predict the future length of hospital stay of new patients,which will help the hospital administration estimate and manage their resources efficiently.

Trend Autoregressive Model Exact Run Length Evaluation on a Two-Sided Extended EWMA Chart

The Extended Exponentially Weighted Moving Average(extended EWMA)control chart is one of the control charts and can be used to quickly detect a small shift.The performance of control charts can be evaluated with the average run length(ARL).Due to the deriving explicit formulas for the ARL on a two-sided extended EWMA control chart for trend autoregressive or trend AR(p)model has not been reported previously.The aim of this study is to derive the explicit formulas for the ARL on a two-sided extended EWMA con-trol chart for the trend AR(p)model as well as the trend AR(1)and trend AR(2)models with exponential white noise.The analytical solution accuracy was obtained with the extended EWMA control chart and was compared to the numer-ical integral equation(NIE)method.The results show that the ARL obtained by the explicit formula and the NIE method is hardly different,but the explicit for-mula can help decrease the computational(CPU)time.Furthermore,this is also expanded to comparative performance with the Exponentially Weighted Moving Average(EWMA)control chart.The performance of the extended EWMA control chart is better than the EWMA control chart for all situations,both the trend AR(1)and trend AR(2)models.Finally,the analytical solution of ARL is applied to real-world data in the healthfield,such as COVID-19 data in the United Kingdom and Sweden,to demonstrate the efficacy of the proposed method.

Analysis of Length of Stay (LOS) Data from the Medical Records of Tertiary Care Hospital in Saudi Arabia for Five Diagnosis Related Groups: Application of Cox Prediction Model

Background: One of the main objectives of hospital managements is to control the length of stay (LOS). Successful control of LOS of inpatients will result in reduction in the cost of care, decrease in nosocomial infections, medication side effects, and better management of the limited number of available patients’ beds. The length of stay (LOS) is an important indicator of the efficiency of hospital management by improving the quality of treatment, and increased hospital profit with more efficient bed management. The purpose of this study was to model the distribution of LOS as a function of patient’s age, and the Diagnosis Related Groups (DRG), based on electronic medical records of a large tertiary care hospital. Materials and Methods: Information related to the research subjects were retrieved from a database of patients admitted to King Faisal Specialist Hospital and Research Center hospital in Riyadh, Saudi Arabia between January 2014 and December 2016. Subjects’ confidential information was masked from the investigators. The data analyses were reported visually, descriptively, and analytically using Cox proportional hazard regression model to predict the risk of long-stay when patients’ age and the DRG are considered as antecedent risk factors. Results: Predicting the risk of long stay depends significantly on the age at admission, and the DRG to which a patient belongs to. We demonstrated the validity of the Cox regression model for the available data as the proportionality assumption is shown to be satisfied. Two examples were presented to demonstrate the utility of the Cox model in this regard.

An Econometric Analysis of Hospital Length of Stay for Cataract Operations in Japan by the Box-Cox Transformation Model and Hausman Tests: Evaluation of the 2010 Revision of the Medical Payment System

The Japanese medical costs for cataract treatments reached 270 billion yen in fiscal year 2012. Since the length of stay (LOS) in hospital is much longer than other major countries, controlling the medical costs by reducing LOS becomes an important issue in Japan. In this paper, we evaluated the effects of the 2010 revision of the Japanese medical payment system (DPC/PDPS) on LOS for cataract operations. The Box-Cox transformation model, Nawata’s estimators and Hausman tests were used in the analysis. To evaluate the effects, we analyzed a dataset obtained from 34 DPC hospitals (Hp1-34) where one-eye cataract operations were performed both before (April 2008-March 2010) and after (April 2010-March 2012) the 2010 revision and there were more than 500 patients. The dataset contained information from 32,593 patients. We did not admit the effect of the 2010 revision in this study, and there were large differences LOS among hospitals, even after removing the influences of factors such as patient characteristics and types of principal diseases.

Proportioning of Cement-Based Grout for Sealing Fractured Rock-Use of Packing Models

Fractured, very permeable rock hosting repositories for radioactive waste will require grouting. New grout types of possible use where long-term performance is needed should have a small amount of cement for minimizing the increase in porosity that will follow from the ultimate dissolution and erosion of this component. They have to be low-viscous and gain strength early after injection and packing theory can assist designers in selecting suitable proportions of various grout components. Optimum particle packing means that the porosity is at minimum and that the amount of cement paste needed to fill the voids between aggregate particles is very small. Low porosity and microstructural stability must be guaranteed for long periods of time. Organic additives for reaching high fluidity cannot be used since they can give off colloids that carry released radionuclides and talc can be an alternative superplasticizer. Low-pH cement reacts with talc to give high strength with time while Portland cement gives early but limited strengthening. The clay mineral palygorskite can be used for early gelation because of its thixotropic properties. Once forced into the rock fractures or channels in soil it stiffens and serves as a filter that prevents fine particles to migrate through it be lost. However, its hydrophilic potential is too high to give the grout a high density and high strength. According to the experiments carried out, most of the investigated grouts are injectable in fractures with apertures down to 100 μm.

Similarities and differences in inherent mechanism and characteristic frequency between the one-dimensional poroelastic model and the layered White model

The similarities and differences in inherent mechanism and characteristic frequency between the onedimensional(1D)poroelastic model and the layered White model were investigated.This investigation was conducted under the assumption that the rock was homogenous and isotropic at the mesoscopic scale.For the inherent mechanism,both models resulted from quasi-static flow in a slow P-wave diffusion mode,and the differences between them originated from saturated fluids and boundary conditions.On the other hand,for the characteristic frequencies of the models,the characteristic frequency of the 1D poroelastic model was first modified because the elastic constant and formula for calculating it were misused and then compared to that of the layered White model.Both of them moved towards higher frequencies with increasing permeability and decreasing viscosity and diffusion length.The differences between them were due to the diffusion length.The diffusion length for the 1D poroelastic model was determined by the sample length,whereas that for the layered White model was determined by the length of the representative elementary volume(REV).Subsequently,a numerical example was presented to demonstrate the similarities and differences between the models.Finally,published experimental data were interpreted using the 1D poroelastic model combined with the Cole-Cole model.The prediction of the combined model was in good agreement with the experimental data,thereby validating the effectiveness of the 1D poroelastic model.Furthermore,the modified characteristic frequency in our study was much closer to the experimental data than the previous prediction,validating the effectiveness of our modification of the characteristic frequency of the 1D poroelastic model.The investigation provided insight into the internal relationship between wave-induced fluid flow(WIFF)models at macroscopic and mesoscopic scales and can aid in a better understanding of the elastic modulus dispersion and attenuation caused by the WIFF at different scales.