On-line least squares support vector machine algorithm in gas prediction

Traditional coal mine safety prediction methods are off-line and do not have dynamic prediction functions.The Support Vector Machine(SVM) is a new machine learning algorithm that has excellent properties.The least squares support vector machine(LS-SVM) algorithm is an improved algorithm of SVM.But the common LS-SVM algorithm,used directly in safety predictions,has some problems.We have first studied gas prediction problems and the basic theory of LS-SVM.Given these problems,we have investigated the affect of the time factor about safety prediction and present an on-line prediction algorithm,based on LS-SVM.Finally,given our observed data,we used the on-line algorithm to predict gas emissions and used other related algorithm to compare its performance.The simulation results have verified the validity of the new algorithm.

Prediction of flare-ups of ulcerative colitis using quantitative immunochemical fecal occult blood test

AIM:To examine the feasibility of predicting the flareup of ulcerative colitis (UC) before symptoms emerge using the immunochemical fecal occult blood test (IFOBT).METHODS:We prospectively measured fecal hemoglobin concentrations in 78 UC patients using the I-FOBT every 1 or 2 mo.RESULTS:During a 20 mo-period,823 fecal samples from 78 patients were submitted.The median concentration of fecal hemoglobin was 41 ng/mL (range:0-392 500 ng/mL).There were three types of patients with regard to the correlation between I-FOBT and patient symptoms;the synchronous transition type with symptoms (44 patients),the unrelated type withsymptoms (19 patients),and the flare-up predictive type (15 patients).In patients with the flare-up predictive type,the values of I-FOBT were generally low during the study period with stable symptoms.Two to four weeks before the flare-up of symptoms,the I-FOBT values were high.Thus,in these patients,I-FOBT could predict the flare-up before symptoms emerged.CONCLUSION:Flare-up could be predicted by I-FOBT in approximately 20% of UC patients.These results warrant periodical I-FOBT in UC patients.

Improving Prediction Efficiency of Machine Learning Models for Cardiovascular Disease in IoST-Based Systems through Hyperparameter Optimization

This study explores the impact of hyperparameter optimization on machine learning models for predicting cardiovascular disease using data from an IoST(Internet of Sensing Things)device.Ten distinct machine learning approaches were implemented and systematically evaluated before and after hyperparameter tuning.Significant improvements were observed across various models,with SVM and Neural Networks consistently showing enhanced performance metrics such as F1-Score,recall,and precision.The study underscores the critical role of tailored hyperparameter tuning in optimizing these models,revealing diverse outcomes among algorithms.Decision Trees and Random Forests exhibited stable performance throughout the evaluation.While enhancing accuracy,hyperparameter optimization also led to increased execution time.Visual representations and comprehensive results support the findings,confirming the hypothesis that optimizing parameters can effectively enhance predictive capabilities in cardiovascular disease.This research contributes to advancing the understanding and application of machine learning in healthcare,particularly in improving predictive accuracy for cardiovascular disease management and intervention strategies.

Prediction method of restoring force based on online AdaBoost regression tree algorithm in hybrid test

In order to solve the poor generalization ability of the back-propagation(BP)neural network in the model updating hybrid test,a novel method called the AdaBoost regression tree algorithm is introduced into the model updating procedure in hybrid tests.During the learning phase,the regression tree is selected as a weak regression model to be trained,and then multiple trained weak regression models are integrated into a strong regression model.Finally,the training results are generated through voting by all the selected regression models.A 2-DOF nonlinear structure was numerically simulated by utilizing the online AdaBoost regression tree algorithm and the BP neural network algorithm as a contrast.The results show that the prediction accuracy of the online AdaBoost regression algorithm is 48.3%higher than that of the BP neural network algorithm,which verifies that the online AdaBoost regression tree algorithm has better generalization ability compared to the BP neural network algorithm.Furthermore,it can effectively eliminate the influence of weight initialization and improve the prediction accuracy of the restoring force in hybrid tests.

Prediction of wax precipitation region in wellbore during deep water oil well testing

During deep water oil well testing, the low temperature environment is easy to cause wax precipitation, which affects the normal operation of the test and increases operating costs and risks. Therefore, a numerical method for predicting the wax precipitation region in oil strings was proposed based on the temperature and pressure fields of deep water test string and the wax precipitation calculation model. And the factors affecting the wax precipitation region were analyzed. The results show that: the wax precipitation region decreases with the increase of production rate, and increases with the decrease of geothermal gradient, increase of water depth and drop of water-cut of produced fluid, and increases slightly with the increase of formation pressure. Due to the effect of temperature and pressure fields, wax precipitation region is large in test strings at the beginning of well production. Wax precipitation region gradually increases with the increase of shut-in time. These conclusions can guide wax prevention during the testing of deep water oil well, to ensure the success of the test.

Comparing Data Mining Techniques in HIV Testing Prediction

Introduction: The present work compared the prediction power of the different data mining techniques used to develop the HIV testing prediction model. Four popular data mining algorithms (Decision tree, Naive Bayes, Neural network, logistic regression) were used to build the model that predicts whether an individual was being tested for HIV among adults in Ethiopia using EDHS 2011. The final experimentation results indicated that the decision tree (random tree algorithm) performed the best with accuracy of 96%, the decision tree induction method (J48) came out to be the second best with a classification accuracy of 79%, followed by neural network (78%). Logistic regression has also achieved the least classification accuracy of 74%. Objectives: The objective of this study is to compare the prediction power of the different data mining techniques used to develop the HIV testing prediction model. Methods: Cross-Industry Standard Process for Data Mining (CRISP-DM) was used to predict the model for HIV testing and explore association rules between HIV testing and the selected attributes. Data preprocessing was performed and missing values for the categorical variable were replaced by the modal value of the variable. Different data mining techniques were used to build the predictive model. Results: The target dataset contained 30,625 study participants. Out of which 16,515 (54%) participants were women while the rest 14,110 (46%) were men. The age of the participants in the dataset ranged from 15 to 59 years old with modal age of 15 - 19 years old. Among the study participants, 17,719 (58%) have never been tested for HIV while the rest 12,906 (42%) had been tested. Residence, educational level, wealth index, HIV related stigma, knowledge related to HIV, region, age group, risky sexual behaviour attributes, knowledge about where to test for HIV and knowledge on family planning through mass media were found to be predictors for HIV testing. Conclusion and Recommendation: The results obtained from this research reveal that data mining is crucial in extracting relevant information for the effective utilization of HIV testing services which has clinical, community and public health importance at all levels. It is vital to apply different data mining techniques for the same settings and compare the model performances (based on accuracy, sensitivity, and specificity) with each other. Furthermore, this study would also invite interested researchers to explore more on the application of data mining techniques in healthcare industry or else in related and similar settings for the future.

Modification of the linear viscoelastic deformation prediction model of asphalt mixture

A deformation prediction model for the dynamic creep test is deduced based on the linear viscoelastic（LVE）theory.Then,the defect of the LVE deformation prediction model is analyzed by comparing the prediction of the LVE deformation model with the experimental data.To improve accuracy,a modification of the LVE deformation prediction model is made to simulate the nonlinear property of the deformation of asphalt mixtures,and it is verified by comparing its simulation results with the experimental data.The comparison results show that the LVE deformation prediction model cannot simulate the nonlinear property of the permanent deformation of asphalt mixtures,while the modified deformation prediction model can provide more precise simulations of the whole process of the deformation and the permanent deformation in the dynamic creep test.Thus,the proposed modification greatly improves the accuracy of the LVE deformation prediction model.The modified model can provide a better understanding of the rutting behavior of asphalt pavement.

Residual lifetime prediction model of nonlinear accelerated degradation data with measurement error

For the product degradation process with random effect (RE), measurement error (ME) and nonlinearity in step-stress accelerated degradation test (SSADT), the nonlinear Wiener based degradation model with RE and ME is built. An analytical approximation to the probability density function (PDF) of the product's lifetime is derived in a closed form. The process and data of SSADT are analyzed to obtain the relation model of the observed data under each accelerated stress. The likelihood function for the population-based observed data is constructed. The population-based model parameters and its random coefficient prior values are estimated. According to the newly observed data of the target product in SSADT, an analytical approximation to the PDF of its residual lifetime (RL) is derived in accordance with its individual degradation characteristics. The parameter updating method based on Bayesian inference is applied to obtain the posterior value of random coefficient of the RL model. A numerical example by simulation is analyzed to verify the accuracy and advantage of the proposed model.

APPLICATION OF WAVELET TRANSFORM ON DIAGNOSIS AND PREDICTION OF MILLING CHATTER

In order to avoid the accuracy deterioration or tool damage caused by milling chatter, it is necessary to have an efficient and reliable diagnosis system that can on-line predict/detect the occur-rence of chatter. The diagnosis/predicting system proposed is to on-line process and analysis the vi-bration signals of the milling machine measured by accelerometers. According to the analysis results, the system will be able to detect/predict the occurrence of the chatter. The diagnosis algorithm is, first, collecting both the normal signals and chatter signals from milling processes, and then, converting the signals through wavelet transform and fast Fourier transform （FFT）. Since the converted chatter sig-nals exhibit different characteristics from the normal signals, through defining the characteristic val-ues, such as root-mean-square value, max value, and ratio of peak value to root-mean-square value, etc, a diagnosis reference library that contains the distribution of these characteristic values is built for diagnosis. When a diagnosis is executing, the characteristic value of the measured signals is con-trasted with the diagnosis reference. The approach index which shows the possibility of occurrence of milling chatter will, then, be calculated through the diagnosis system. Cutting experiments are con-ducted to verify the proposed diagnosis system. The results show the success of early chatter detecting for the system.

Predictive value of hypo-osmotic swelling test to identifyviable non-motile sperm

Aim: To determine the predictive value of the hypo-osmotic swelling (HOS) test to identify viable, non-motile sperm. Methods: Semen samples from 20 men with severe asthenozoospermia underwent traditional seminal analysis, eosin-nigrosin (EN) staining and the HOS test. A further EN stain was then performed on a HOS pre-treated aliquot and a total of 2000 further sperm examined. Results: The median sperm density was 5.1 million/mL (IQR 4.3-13.1) and the median motility was 3.0 % (IQR 0-7). Seven samples showed complete asthenozoospermia. Initial EN staining showed 59 % viability (range 48-69) despite the poor standard parameters and 47 % (range 33-61) in the complete asthenozoospermia subgroup. The HOS test showed 49.9 % reacted overall (range 40-59) and 41.7 % (range 22-61) in the complete asthenozoospermia subgroup. The combined HOS/EN stain showed the positive predictive value of the HOS test to identify viable sperm was 84.2 % overall and 79.7 % in the complete asthenozoospermia subgroup. Conclusion: The HOS test can effectively predict sperm viability in patients with severe and complete asthenozoospermia.

Fracture gradient prediction:an overview and an improved method

The fracture gradient is a critical parameter for drilling mud weight design in the energy industry. A new method in fracture gradient prediction is proposed based on analyzing worldwide leak-off test(LOT) data in offshore drilling. Current fracture gradient prediction methods are also reviewed and compared to the proposed method. We analyze more than 200 LOT data in several offshore petroleum basins and find that the fracture gradient depends not only on the overburden stress and pore pressure, but also on the depth. The data indicate that the effective stress coefficient is higher at a shallower depth than that at a deeper depth in the shale formations. Based on this finding,a depth-dependent effective stress coefficient is proposed and applied for fracture gradient prediction. In some petroleum basins, many wells need to be drilled through long sections of salt formations to reach hydrocarbon reservoirs.The fracture gradient in salt formations is very different from that in other sedimentary rocks. Leak-off test data in the salt formations are investigated, and a fracture gradient prediction method is proposed. Case applications are examined to compare different fracture gradient methods and validate the proposed methods. The reasons why the LOT value is higher than its overburden gradient are also explained.

Prediction of recurrence and prognosis in patients with hepatocellular carcinoma after resection by use of CLIP score

AIM: The survival time of patients with hepatocellular carcinoma (HCC) after resection is hard to predict. Both residual liver function and tumor extension factors should be considered. A new scoring system has recently been proposed by the Cancer of the Liver Italian Program (CLIP). CLIP score was confirmed to be one of the best ways to stage patients with HCC. To our knowledge, however, the literature concerning the correlation between CLIP score and prognosis for patients with HCC after resection was not published. The aim of this study is to evaluate the recurrence and prognostic value of CLIP score for the patients with HCC after resection. METHODS: A retrospective survey was carried out in 174 patients undergoing resection of HCC from January 1986 to June 1998. Six patients who died in the hospital after operation and 11 patients with the recurrence of the disease were excluded at 1 month after hepatectomy. By the end of June 2001, 4 patients were lost and 153 patients with curative resection have been followed up for at least three years. Among 153 patients, 115 developed intrahepatic recurrence and 10 developed extrahepatic recurrence, whereas the other 28 remained free of recurrence. Recurrences were classified into early (【 or =3 year) and late (】3 year) recurrence. The CLIP score included the parameters involved in the Child-Pugh stage (0-2), plus macroscopic tumor morphology (0-2), AFP levels (0-1), and the presence or absence of portal thrombosis (0-1). By contrast, portal vein thrombosis was defined as the presence of tumor emboli within vascular channel analyzed by microscopic examination in this study. Risk factors for recurrence and prognostic factors for survival in each group were analyzed by the chi-square test, the Kaplan-Meier estimation and the COX proportional hazards model respectively. RESULTS: The 1-, 3-, 5-, 7-,and 10-year disease-free survival rates after curative resection of HCC were 57.2%, 28.3%, 23.5%, 18.8%, and 17.8%, respectively. Median survival time was 28, 10, 4, and 5 mo for CLIP score 0, 1, 2, 3, and 4 to 5, respectively. Early and late recurrence developed in 109 patients and 16 patients respectively. By the chi-square test, tumor size, microsatellite, venous invasion, tumor type (uninodular, multinodular, massive), tumor extension (【 or = or 】50% of liver parenchyma replaced by tumor), TNM stage, CLIP score, and resection margin were the risk factors for early recurrence, whereas CLIP score and Child-Pugh stage were significant risk factors for late recurrence. In univariate survival analysis, Child-Pugh stages, resection margin, tumor size, microsatellite, venous invasion, tumor type, tumor extension, TNM stages, and CLIP score were associated with prognosis. The multivariate analysis by COX proportional hazards model showed that the independent predictive factors of survival were resection margins and TNM stages. CONCLUSION: CLIP score has displayed a unique superiority in predicting the tumor early and late recurrence and prognosis in the patients with HCC after resection.

Application of quantitative estimates of fecal hemoglobin concentration for risk prediction of colorectal neoplasia

AIM:To determine the role of the fecal immunochemical test(FIT),used to evaluate fecal hemoglobin concentration,in the prediction of histological grade and risk of colorectal tumors.METHODS:We enrolled 17881 individuals who attended the two-step colorectal cancer screening program in a single hospital between January 2010 and October 2011.Colonoscopy was recommended to the participants with an FIT of≥12 ngHb/mL buffer.We classified colorectal lesions as cancer(C),advanced adenoma(AA),adenoma(A),and others(O)by their colonoscopic and histological findings.Multiple linear regression analysis adjusted for age and gender was used to determine the association between the FIT results and colorectal tumor grade.The risk of adenomatous neoplasia was estimated by calculating the positive predictive values for different FIT concentrations.RESULTS:The positive rate of the FIT was 10.9%(1948/17881).The attendance rate for colonoscopy was 63.1%(1229/1948).The number of false positive results was 23.Of these 1229 cases,the numbers of O,A,AA,and C were 759,221,201,and 48,respectively.Regression analysis revealed a positive association between histological grade and FIT concentration(β=0.088,P<0.01).A significant log-linear relationship was found between the concentration and positive predictive value of the FIT for predicting colorectal tumors(R2>0.95,P<0.001).CONCLUSION:Higher FIT concentrations are associated with more advanced histological grades.Risk prediction for colorectal neoplasia based on individual FIT concentrations is significant and may help to improve the performance of screening programs.

A male-ABCD algorithm for hepatocellular carcinoma risk prediction in HBs Ag carriers

Objective: Hepatocellular carcinoma(HCC) development among hepatitis B surface antigen(HBs Ag) carriers shows gender disparity, influenced by underlying liver diseases that display variations in laboratory tests. We aimed to construct a risk-stratified HCC prediction model for HBs Ag-positive male adults.Methods: HBs Ag-positive males of 35-69 years old(N=6,153) were included from a multi-center populationbased liver cancer screening study. Randomly, three centers were set as training, the other three centers as validation. Within 2 years since initiation, we administrated at least two rounds of HCC screening using Bultrasonography and α-fetoprotein(AFP). We used logistic regression models to determine potential risk factors,built and examined the operating characteristics of a point-based algorithm for HCC risk prediction.Results: With 2 years of follow-up, 302 HCC cases were diagnosed. A male-ABCD algorithm was constructed including participant's age, blood levels of GGT(γ-glutamyl-transpeptidase), counts of platelets, white cells,concentration of DCP(des-γ-carboxy-prothrombin) and AFP, with scores ranging from 0 to 18.3. The area under receiver operating characteristic was 0.91(0.90-0.93), larger than existing models. At 1.5 points of risk score,26.10% of the participants in training cohort and 14.94% in validation cohort were recognized at low risk, with sensitivity of identifying HCC remained 100%. At 2.5 points, 46.51% of the participants in training cohort and 33.68% in validation cohort were recognized at low risk with 99.06% and 97.78% of sensitivity, respectively. At 4.5 points, only 20.86% of participants in training cohort and 23.73% in validation cohort were recognized at high risk,with positive prediction value of 22.85% and 12.35%, respectively.Conclusions: Male-ABCD algorithm identified individual's risk for HCC occurrence within short term for their HCC precision surveillance.

Prediction of mechanical properties for deep drawing steel by deep learning

At present,iron and steel enterprises mainly use“after spot test ward”to control final product quality.However,it is impossible to realize on-line quality predetermining for all products by this traditional approach,hence claims and returns often occur,resulting in major eco-nomic losses of enterprises.In order to realize the on-line quality predetermining for steel products during manufacturing process,the predic-tion models of mechanical properties based on deep learning have been proposed in this work.First,the mechanical properties of deep drawing steels were predicted by using LSTM(long short team memory),GRU(gated recurrent unit)network,and GPR(Gaussian process regression)model,and prediction accuracy and learning efficiency for different models were also discussed.Then,on-line re-learning methods for transfer learning models and model parameters were proposed.The experimental results show that not only the prediction accuracy of optimized trans-fer learning models has been improved,but also predetermining time was shortened to meet real time requirements of on-line property prede-termining.The industrial production data of interstitial-free(IF)steel was used to demonstrate that R2 value of GRU model in training stage reaches more than 0.99,and R2 value in testing stage is more than 0.96.

Quality prediction of freshly-harvested wheat using Gluto Peak during postharvest maturation

Much recent researches have demonstrated that the quality of freshly-harvested wheat could be improved during postharvest maturation by determinating the rheological properties.However,this process is time-consuming and complex.This study aimed to provide a rapid and convenient method for predicting wheat quality during postharvest maturation by use of Gluto Peak device.Farinograph and Extensograph were used to determine the rheological properties of four wheat samples(WT1,WT2,WT3,WT4)stored under different conditions(WT1:15℃,50%RH;WT2:20℃65%RH;WT3:28℃75%RH;WT4:35℃85%RH)for a total of 10 weeks,and Gluto Peak test was used to determine the gluten aggregation properties of the four samples.Correlation analysis was also conducted between the rheological properties and the gluten aggregation properties.Results of rheological properties showed that all Extensographic properties(dough extensibility,resistance,maximum resistance and area)of the four samples increased along with the storage time,and the Farinographic properties(water absorption,dough development time,dough stability time,and farinograph quality number(FQN))had the same tendency,indicating that the rheological properties were improved considerably with storage time extending.The Gluto Peak curves revealed that Peak Maximum Time(PMT),Brabender Equivalents Maximum(BEM)and Energy to Maximum Torque(En MT)of wheat flour of the four samples varied greatly,particularly the PMT and En MT of the samples WT3 and WT4 increased remarkably.Results of correlation analysis showed that En MT had significant correlation with water absorption and area(P<0.05)for sample WT1,and also showed significant correlation with dough development time(P<0.05)for sample WT2.For sample WT3,PMT was significantly correlated with the dough development time,extensibility,area(P<0.05),and FQN(P<0.01);and En MT was in significant correction with water absorption(P<0.01),and dough stability time,FQN,extensibility,maximum resistance and area(P<0.05).For sample WT4,both PMT and En MT had significant correction with area(P<0.05).The study indicated that the Gluto Peak test is effective in quality prediction for the freshly-harvested wheat during postharvest maturation,making it possible to realize rapid wheat quality detection and evaluation in storage period.

On-line real-time path planning of mobile robots in dynamic uncertain environment

A new path planning method for mobile robots in globally unknown environment with moving obstacles is pre- sented. With an autoregressive (AR) model to predict the future positions of moving obstacles, and the predicted position taken as the next position of moving obstacles, a motion path in dynamic uncertain environment is planned by means of an on-line real-time path planning technique based on polar coordinates in which the desirable direction angle is taken into consideration as an optimization index. The effectiveness, feasibility, high stability, perfect performance of obstacle avoidance, real-time and optimization capability are demonstrated by simulation examples.

The RCT Approach for Cell Coverage Prediction (Ⅱ): Experiment and Early Results

We study on the implementation flow of the radio computerized tomography (RCT) prediction method. A case in real cellular mobile radio (CMR) system together with the prediction results are also presented. As shown by the results, the RCT prediction method is marked for its convenience and rapidity, as well as its relative high precision even when the prediction procedure is highly simplified. Since it is developed according to the characteristics of wireless communication environments of our country and has concurrently merits from both statistical and deterministic prediction models, the RCT prediction method is in good agreement with engineering practices in cellular mobile communication in cities at home. Optimized by combining with other techniques, further improvement could be achieved in the stability and precision of the RCT prediction method which now serves as the core part of a software tool for commercial use in CMR system analysis and optimization.

Modeling of Microstructure Evolution and Mechanical Properties of Steel Plates Produced by Thermo-Mechanical Control Process and Its On-line Application

An integrated metallurgical model was developed to predict microstructure evolution and mechanical properties of low-carbon steel plates produced by TMCP. The metallurgical phenomena occurring during TMCP and mechanical properties were predicted for different process parameters. In the later passes full recrystallization becomes difficult to occur and higher residual strain remains in austenite after rolling. For the reasonable temperature and cooling schedule, yield strength of 30 mm plain carbon steel plate can reach 310 MPa. The first on-line application of prediction and control of microstructure and properties (PCMP) in the medium plate production was achieved. The predictions of the system are in good agreement with measurements.