Any disease can be treated only once it is imaged,detected and classified.This paper proposes a set of algorithms for classification of a brain tumor with better accuracy and efficiency.The proposal uses a JPEG format...Any disease can be treated only once it is imaged,detected and classified.This paper proposes a set of algorithms for classification of a brain tumor with better accuracy and efficiency.The proposal uses a JPEG format of the DICOM image fed into three stages namely pre-processing,segmentation using sub-pixel edge detection method and using the nearest neighbor methodology for the detection and differentiation of benign and malignant tumors.展开更多
The Tayatea Dyke Swarm(also known as the Tayatea Dolerite)comprises well-exposed northeast-trending tholeiitic dykes that intrude the Rocky Cape Group(RCG)of northwest Tasmania,Australia.The dykes commonly
1st cases of COVID-19 were reported in March 2020 in Bangladesh and rapidly increased daily. So many steps were taken by the Bangladesh government to reduce the outbreak of COVID-19, such as masks, gatherings, local m...1st cases of COVID-19 were reported in March 2020 in Bangladesh and rapidly increased daily. So many steps were taken by the Bangladesh government to reduce the outbreak of COVID-19, such as masks, gatherings, local movements, international movements, etc. The data was collected from the World Health Organization. In this research, different variables have been used for analysis, for instance, new cases, new deaths, masks, schools, business, gatherings, domestic movement, international travel, new test, positive rate, test per case, new vaccination smoothed, new vaccine, total vaccination, and stringency index. Machine learning algorithms were used to predict and build the model, such as linear regression, K-nearest neighbours, decision trees, random forests, and support vector machines. Accuracy and Mean Square error (MSE) were used to test the model. A hyperparameter was also applied to find the optimum values of parameters. After computing the analysis, the result showed that the linear regression algorithm performs the best overall among the algorithms listed, with the highest testing accuracy and the lowest RMSE before and after hyper-tuning. The highest accuracy and lowest MSE were used for the best model, and for this data set, Linear regression got the highest accuracy, 0.98 and 0.97 and the lowest MSE, 4.79 and 4.04, respectively.展开更多
BACKGROUND Severe dengue children with critical complications have been attributed to high mortality rates,varying from approximately 1%to over 20%.To date,there is a lack of data on machine-learning-based algorithms ...BACKGROUND Severe dengue children with critical complications have been attributed to high mortality rates,varying from approximately 1%to over 20%.To date,there is a lack of data on machine-learning-based algorithms for predicting the risk of inhospital mortality in children with dengue shock syndrome(DSS).AIM To develop machine-learning models to estimate the risk of death in hospitalized children with DSS.METHODS This single-center retrospective study was conducted at tertiary Children’s Hospital No.2 in Viet Nam,between 2013 and 2022.The primary outcome was the in-hospital mortality rate in children with DSS admitted to the pediatric intensive care unit(PICU).Nine significant features were predetermined for further analysis using machine learning models.An oversampling method was used to enhance the model performance.Supervised models,including logistic regression,Naïve Bayes,Random Forest(RF),K-nearest neighbors,Decision Tree and Extreme Gradient Boosting(XGBoost),were employed to develop predictive models.The Shapley Additive Explanation was used to determine the degree of contribution of the features.RESULTS In total,1278 PICU-admitted children with complete data were included in the analysis.The median patient age was 8.1 years(interquartile range:5.4-10.7).Thirty-nine patients(3%)died.The RF and XGboost models demonstrated the highest performance.The Shapley Addictive Explanations model revealed that the most important predictive features included younger age,female patients,presence of underlying diseases,severe transaminitis,severe bleeding,low platelet counts requiring platelet transfusion,elevated levels of international normalized ratio,blood lactate and serum creatinine,large volume of resuscitation fluid and a high vasoactive inotropic score(>30).CONCLUSION We developed robust machine learning-based models to estimate the risk of death in hospitalized children with DSS.The study findings are applicable to the design of management schemes to enhance survival outcomes of patients with DSS.展开更多
By using the method of density-matrix renormalization-group to solve the different spin spin correlation functions, the nearest-neighbouring entanglement （NNE） and the next-nearest-neighbouring entanglement （NNNE）...By using the method of density-matrix renormalization-group to solve the different spin spin correlation functions, the nearest-neighbouring entanglement （NNE） and the next-nearest-neighbouring entanglement （NNNE） of one-dimensional alternating Heisenberg XY spin chain are investigated in the presence of alternating the-nearestneighbouring interaction of exchange couplings, external magnetic fields and the next-nearest neighbouring interaction. For a dimerised ferromagnetic spin chain, the NNNE appears only above a critical dimerized interaction, meanwhile, the dimerized interaction a effects a quantum phase transition point and improves the NNNE to a large extent. We also study the effect of ferromagnetic or antiferromagnetic next-nearest neighbouring （NNN） interaction on the dynamics of NNE and NNNE. The ferromagnetic NNN interaction increases and shrinks the NNE below and above a critical frustrated interaction respectively, while the antiferromagnetic NNN interaction always reduces the NNE. The antiferromagnetic NNN interaction results in a large value of NNNE compared with the case where the NNN interaction is ferromagnetic.展开更多
The paper systematically deals with the background of regional isotopic compo-sitions in the lower and middle reaches of the Yangtze River and neighbouring areas. It isshown that the lead isotopic compositions of diff...The paper systematically deals with the background of regional isotopic compo-sitions in the lower and middle reaches of the Yangtze River and neighbouring areas. It isshown that the lead isotopic compositions of different geological formations and units are con-trolled by the primary mantle heterogeneity, dynamic process of crust-mantle interchange,abundances of uraninm, thorium and lead of various layers of the earth and timing. Studies onthe background of regional isotopic compositions may offer significant information forgeochemical regionalization, tracing of sources of ore-forming materials, and regionalprognosis of ore deposits.展开更多
Arrhythmia beat classification is an active area of research in ECG based clinical decision support systems. In this paper, Pruned Fuzzy K-nearest neighbor (PFKNN) classifier is proposed to classify six types of beats...Arrhythmia beat classification is an active area of research in ECG based clinical decision support systems. In this paper, Pruned Fuzzy K-nearest neighbor (PFKNN) classifier is proposed to classify six types of beats present in the MIT-BIH Arrhythmia database. We have tested our classifier on ~ 103100 beats for six beat types present in the database. Fuzzy KNN (FKNN) can be implemented very easily but large number of training examples used for classification can be very time consuming and requires large storage space. Hence, we have proposed a time efficient Arif-Fayyaz pruning algorithm especially suitable for FKNN which can maintain good classification accuracy with appropriate retained ratio of training data. By using Arif-Fayyaz pruning algorithm with Fuzzy KNN, we have achieved a beat classification accuracy of 97% and geometric mean of sensitivity of 94.5% with only 19% of the total training examples. The accuracy and sensitivity is comparable to FKNN when all the training data is used. Principal Component Analysis is used to further reduce the dimension of feature space from eleven to six without compromising the accuracy and sensitivity. PFKNN was found to robust against noise present in the ECG data.展开更多
During the storehouse surface rolling construction of a core rockfilldam, the spreading thickness of dam face is an important factor that affects the construction quality of the dam storehouse' rolling surface and...During the storehouse surface rolling construction of a core rockfilldam, the spreading thickness of dam face is an important factor that affects the construction quality of the dam storehouse' rolling surface and the overallquality of the entire dam. Currently, the method used to monitor and controlspreading thickness during the dam construction process is artificialsampling check after spreading, which makes it difficult to monitor the entire dam storehouse surface. In this paper, we present an in-depth study based on real-time monitoring and controltheory of storehouse surface rolling construction and obtain the rolling compaction thickness by analyzing the construction track of the rolling machine. Comparatively, the traditionalmethod can only analyze the rolling thickness of the dam storehouse surface after it has been compacted and cannot determine the thickness of the dam storehouse surface in realtime. To solve these problems, our system monitors the construction progress of the leveling machine and employs a real-time spreading thickness monitoring modelbased on the K-nearest neighbor algorithm. Taking the LHK core rockfilldam in Southwest China as an example, we performed real-time monitoring for the spreading thickness and conducted real-time interactive queries regarding the spreading thickness. This approach provides a new method for controlling the spreading thickness of the core rockfilldam storehouse surface.展开更多
Short-term traffic flow is one of the core technologies to realize traffic flow guidance. In this article, in view of the characteristics that the traffic flow changes repeatedly, a short-term traffic flow forecasting...Short-term traffic flow is one of the core technologies to realize traffic flow guidance. In this article, in view of the characteristics that the traffic flow changes repeatedly, a short-term traffic flow forecasting method based on a three-layer K-nearest neighbor non-parametric regression algorithm is proposed. Specifically, two screening layers based on shape similarity were introduced in K-nearest neighbor non-parametric regression method, and the forecasting results were output using the weighted averaging on the reciprocal values of the shape similarity distances and the most-similar-point distance adjustment method. According to the experimental results, the proposed algorithm has improved the predictive ability of the traditional K-nearest neighbor non-parametric regression method, and greatly enhanced the accuracy and real-time performance of short-term traffic flow forecasting.展开更多
Background:Species-specific genotypic features,local neighbourhood interactions and resource supply strongly influence the tree stature and growth rate.In mixed-species forests,diversity-mediated biomass allocation ha...Background:Species-specific genotypic features,local neighbourhood interactions and resource supply strongly influence the tree stature and growth rate.In mixed-species forests,diversity-mediated biomass allocation has been suggested to be a fundamental mechanism underlying the positive biodiversity-productivity relationships.Empirical evidence,however,is rare about the impact of local neighbourhood diversity on tree characteristics analysed at a very high level of detail.To address this issue we analysed these effects on the individual-tree crown architecture and tree productivity in a mature mixed forest in northern Germany.Methods:Our analysis considers multiple target tree species across a local neighbourhood species richness gradient ranging from 1 to 4.We applied terrestrial laser scanning to quantify a large number of individual mature trees(N=920)at very high accuracy.We evaluated two different neighbour inclusion approaches by analysing both a fixed radius selection procedure and a selection based on overlapping crowns.Results and conclusions:We show that local neighbourhood species diversity significantly increases crown dimension and wood volume of target trees.Moreover,we found a size-dependency of diversity effects on tree productivity(basal area and wood volume increment)with positive effects for large-sized trees(diameter at breast height(DBH)>40 cm)and negative effects for small-sized(DBH<40 cm)trees.In our analysis,the neighbour inclusion approach has a significant impact on the outcome.For scientific studies and the validation of growth models we recommend a neighbour selection by overlapping crowns,because this seems to be the relevant scale at which local neighbourhood interactions occur.Because local neighbourhood diversity promotes individual-tree productivity in mature European mixed-species forests,we conclude that a small-scale species mixture should be considered in management plans.展开更多
As climate change negotiations progress,monitoring biomass and carbon stocks is becoming an important part of the current forest research.Therefore,national governments are interested in developing forest-monitoring s...As climate change negotiations progress,monitoring biomass and carbon stocks is becoming an important part of the current forest research.Therefore,national governments are interested in developing forest-monitoring strategies using geospatial technology.Among statistical methods for mapping biomass,there is a nonparametric approach called k-nearest neighbor(kNN).We compared four variations of distance metrics of the kNN for the spatially-explicit estimation of aboveground biomass in a portion of the Mexican north border of the intertropical zone.Satellite derived,climatic,and topographic predictor variables were combined with the Mexican National Forest Inventory(NFI)data to accomplish the purpose.Performance of distance metrics applied into the kNN algorithm was evaluated using a cross validation leave-one-out technique.The results indicate that the Most Similar Neighbor(MSN)approach maximizes the correlation between predictor and response variables(r=0.9).Our results are in agreement with those reported in the literature.These findings confirm the predictive potential of the MSN approach for mapping forest variables at pixel level under the policy of Reducing Emission from Deforestation and Forest Degradation(REDD+).展开更多
On the basis of machine leaning,suitable algorithms can make advanced time series analysis.This paper proposes a complex k-nearest neighbor(KNN)model for predicting financial time series.This model uses a complex feat...On the basis of machine leaning,suitable algorithms can make advanced time series analysis.This paper proposes a complex k-nearest neighbor(KNN)model for predicting financial time series.This model uses a complex feature extraction process integrating a forward rolling empirical mode decomposition(EMD)for financial time series signal analysis and principal component analysis(PCA)for the dimension reduction.The information-rich features are extracted then input to a weighted KNN classifier where the features are weighted with PCA loading.Finally,prediction is generated via regression on the selected nearest neighbors.The structure of the model as a whole is original.The test results on real historical data sets confirm the effectiveness of the models for predicting the Chinese stock index,an individual stock,and the EUR/USD exchange rate.展开更多
The natural neighbour method can be considered as one of many variants of the meshless methods. In the present paper, a new approach based on the Fraeijs de Veubeke （FdV） functional, which is initially developed for...The natural neighbour method can be considered as one of many variants of the meshless methods. In the present paper, a new approach based on the Fraeijs de Veubeke （FdV） functional, which is initially developed for linear elasticity, is extended to the case of geometrically linear but materially non-linear solids. The new approach provides an original treatment to two classical problems： the numerical evaluation of the integrals over the domain A and the enforcement of boundary conditions of the type ui = hi on Su. In the absence of body forces （Fi = 0）, it will be shown that the calculation of integrals of the type fA .dA can be avoided and that boundary conditions of the type ui = hi on Su can be imposed in the average sense in general and exactly if hi is linear between two contour nodes, which is obviously the case for tTi = O.展开更多
This paper proposes an active learning accelerated Monte-Carlo simulation method based on the modified K-nearest neighbors algorithm.The core idea of the proposed method is to judge whether or not the output of a rand...This paper proposes an active learning accelerated Monte-Carlo simulation method based on the modified K-nearest neighbors algorithm.The core idea of the proposed method is to judge whether or not the output of a random input point can be postulated through a classifier implemented through the modified K-nearest neighbors algorithm.Compared to other active learning methods resorting to experimental designs,the proposed method is characterized by employing Monte-Carlo simulation for sampling inputs and saving a large portion of the actual evaluations of outputs through an accurate classification,which is applicable for most structural reliability estimation problems.Moreover,the validity,efficiency,and accuracy of the proposed method are demonstrated numerically.In addition,the optimal value of K that maximizes the computational efficiency is studied.Finally,the proposed method is applied to the reliability estimation of the carbon fiber reinforced silicon carbide composite specimens subjected to random displacements,which further validates its practicability.展开更多
In the contemporary era of unprecedented innovations such as Internet of Things(IoT),modern applications cannot be imagined without the presence of Wireless Sensor Network(WSN).Nodes in WSN use neighbour discovery(ND)...In the contemporary era of unprecedented innovations such as Internet of Things(IoT),modern applications cannot be imagined without the presence of Wireless Sensor Network(WSN).Nodes in WSN use neighbour discovery(ND)protocols to have necessary communication among the nodes.Neighbour discovery process is crucial as it is to be done with energy efficiency and minimize discovery latency and maximize percentage of neighbours discovered.The current ND approaches that are indirect in nature are categorized into methods of removal of active slots from wake-up schedules and intelligent addition of new slots.The two methods are found to have certain drawbacks.Thefirst category disturbs original integrity of wake-up schedules leading to reduced chances of discovering new nodes in WSN as neighbours.When second category is followed,it may have inefficient slots in the wake-up schedules leading to performance degradation.Therefore,the motivation behind the work in this paper is that by combining the two categories,it is possible to reap benefits of both and get rid of the limitations of the both.Making a hybrid is achieved by introducing virtual nodes that help maximize performance by ensuring original integrity of wake-up schedules and adding of efficient active slots.Thus a Hybrid Approach to Neighbour Discovery(HAND)protocol is realized in WSN.The simulation study revealed that HAND outperforms the existing indirect ND models.展开更多
Stock trend prediction is a challenging problem because it involves many variables.Aiming at the problem that some existing machine learning techniques, such as random forest(RF), probabilistic random forest(PRF), k-n...Stock trend prediction is a challenging problem because it involves many variables.Aiming at the problem that some existing machine learning techniques, such as random forest(RF), probabilistic random forest(PRF), k-nearest neighbor(KNN), and fuzzy KNN(FKNN), have difficulty in accurately predicting the stock trend(uptrend or downtrend) for a given date, a generalized Heronian mean(GHM) based FKNN predictor named GHM-FKNN was proposed.GHM-FKNN combines GHM aggregation function with the ideas of the classical FKNN approach.After evaluation, the comparison results elucidated that GHM-FKNN outperformed the other best existing methods RF, PRF, KNN and FKNN on independent test datasets corresponding to three stocks, namely AAPL, AMZN and NFLX.Compared with RF, PRF, KNN and FKNN, GHM-FKNN achieved the best performance with accuracy of 62.37% for AAPL, 58.25% for AMZN, and 64.10% for NFLX.展开更多
文摘Any disease can be treated only once it is imaged,detected and classified.This paper proposes a set of algorithms for classification of a brain tumor with better accuracy and efficiency.The proposal uses a JPEG format of the DICOM image fed into three stages namely pre-processing,segmentation using sub-pixel edge detection method and using the nearest neighbor methodology for the detection and differentiation of benign and malignant tumors.
文摘The Tayatea Dyke Swarm(also known as the Tayatea Dolerite)comprises well-exposed northeast-trending tholeiitic dykes that intrude the Rocky Cape Group(RCG)of northwest Tasmania,Australia.The dykes commonly
文摘1st cases of COVID-19 were reported in March 2020 in Bangladesh and rapidly increased daily. So many steps were taken by the Bangladesh government to reduce the outbreak of COVID-19, such as masks, gatherings, local movements, international movements, etc. The data was collected from the World Health Organization. In this research, different variables have been used for analysis, for instance, new cases, new deaths, masks, schools, business, gatherings, domestic movement, international travel, new test, positive rate, test per case, new vaccination smoothed, new vaccine, total vaccination, and stringency index. Machine learning algorithms were used to predict and build the model, such as linear regression, K-nearest neighbours, decision trees, random forests, and support vector machines. Accuracy and Mean Square error (MSE) were used to test the model. A hyperparameter was also applied to find the optimum values of parameters. After computing the analysis, the result showed that the linear regression algorithm performs the best overall among the algorithms listed, with the highest testing accuracy and the lowest RMSE before and after hyper-tuning. The highest accuracy and lowest MSE were used for the best model, and for this data set, Linear regression got the highest accuracy, 0.98 and 0.97 and the lowest MSE, 4.79 and 4.04, respectively.
文摘BACKGROUND Severe dengue children with critical complications have been attributed to high mortality rates,varying from approximately 1%to over 20%.To date,there is a lack of data on machine-learning-based algorithms for predicting the risk of inhospital mortality in children with dengue shock syndrome(DSS).AIM To develop machine-learning models to estimate the risk of death in hospitalized children with DSS.METHODS This single-center retrospective study was conducted at tertiary Children’s Hospital No.2 in Viet Nam,between 2013 and 2022.The primary outcome was the in-hospital mortality rate in children with DSS admitted to the pediatric intensive care unit(PICU).Nine significant features were predetermined for further analysis using machine learning models.An oversampling method was used to enhance the model performance.Supervised models,including logistic regression,Naïve Bayes,Random Forest(RF),K-nearest neighbors,Decision Tree and Extreme Gradient Boosting(XGBoost),were employed to develop predictive models.The Shapley Additive Explanation was used to determine the degree of contribution of the features.RESULTS In total,1278 PICU-admitted children with complete data were included in the analysis.The median patient age was 8.1 years(interquartile range:5.4-10.7).Thirty-nine patients(3%)died.The RF and XGboost models demonstrated the highest performance.The Shapley Addictive Explanations model revealed that the most important predictive features included younger age,female patients,presence of underlying diseases,severe transaminitis,severe bleeding,low platelet counts requiring platelet transfusion,elevated levels of international normalized ratio,blood lactate and serum creatinine,large volume of resuscitation fluid and a high vasoactive inotropic score(>30).CONCLUSION We developed robust machine learning-based models to estimate the risk of death in hospitalized children with DSS.The study findings are applicable to the design of management schemes to enhance survival outcomes of patients with DSS.
基金Project supported by the Key Higher Education Program of Hubei Province, China (Grant No Z20052201)Natural Science Foundation of Hubei Province, China (Grant No 2006ABA055)Postgraduate Program of Hubei Normal University of China(Grant No 2007D20)
文摘By using the method of density-matrix renormalization-group to solve the different spin spin correlation functions, the nearest-neighbouring entanglement （NNE） and the next-nearest-neighbouring entanglement （NNNE） of one-dimensional alternating Heisenberg XY spin chain are investigated in the presence of alternating the-nearestneighbouring interaction of exchange couplings, external magnetic fields and the next-nearest neighbouring interaction. For a dimerised ferromagnetic spin chain, the NNNE appears only above a critical dimerized interaction, meanwhile, the dimerized interaction a effects a quantum phase transition point and improves the NNNE to a large extent. We also study the effect of ferromagnetic or antiferromagnetic next-nearest neighbouring （NNN） interaction on the dynamics of NNE and NNNE. The ferromagnetic NNN interaction increases and shrinks the NNE below and above a critical frustrated interaction respectively, while the antiferromagnetic NNN interaction always reduces the NNE. The antiferromagnetic NNN interaction results in a large value of NNNE compared with the case where the NNN interaction is ferromagnetic.
基金This study was co-supported by the State Eighth Five-Year Plan Scientific Project(No.85-901-03-08D)and National Natural Science Foundation of China(No.49473187).
文摘The paper systematically deals with the background of regional isotopic compo-sitions in the lower and middle reaches of the Yangtze River and neighbouring areas. It isshown that the lead isotopic compositions of different geological formations and units are con-trolled by the primary mantle heterogeneity, dynamic process of crust-mantle interchange,abundances of uraninm, thorium and lead of various layers of the earth and timing. Studies onthe background of regional isotopic compositions may offer significant information forgeochemical regionalization, tracing of sources of ore-forming materials, and regionalprognosis of ore deposits.
文摘Arrhythmia beat classification is an active area of research in ECG based clinical decision support systems. In this paper, Pruned Fuzzy K-nearest neighbor (PFKNN) classifier is proposed to classify six types of beats present in the MIT-BIH Arrhythmia database. We have tested our classifier on ~ 103100 beats for six beat types present in the database. Fuzzy KNN (FKNN) can be implemented very easily but large number of training examples used for classification can be very time consuming and requires large storage space. Hence, we have proposed a time efficient Arif-Fayyaz pruning algorithm especially suitable for FKNN which can maintain good classification accuracy with appropriate retained ratio of training data. By using Arif-Fayyaz pruning algorithm with Fuzzy KNN, we have achieved a beat classification accuracy of 97% and geometric mean of sensitivity of 94.5% with only 19% of the total training examples. The accuracy and sensitivity is comparable to FKNN when all the training data is used. Principal Component Analysis is used to further reduce the dimension of feature space from eleven to six without compromising the accuracy and sensitivity. PFKNN was found to robust against noise present in the ECG data.
基金supported by the Innovative Research Groups of National Natural Science Foundation of China(No. 51621092)National Basic Research Program of China ("973" Program, No. 2013CB035904)National Natural Science Foundation of China (No. 51439005)
文摘During the storehouse surface rolling construction of a core rockfilldam, the spreading thickness of dam face is an important factor that affects the construction quality of the dam storehouse' rolling surface and the overallquality of the entire dam. Currently, the method used to monitor and controlspreading thickness during the dam construction process is artificialsampling check after spreading, which makes it difficult to monitor the entire dam storehouse surface. In this paper, we present an in-depth study based on real-time monitoring and controltheory of storehouse surface rolling construction and obtain the rolling compaction thickness by analyzing the construction track of the rolling machine. Comparatively, the traditionalmethod can only analyze the rolling thickness of the dam storehouse surface after it has been compacted and cannot determine the thickness of the dam storehouse surface in realtime. To solve these problems, our system monitors the construction progress of the leveling machine and employs a real-time spreading thickness monitoring modelbased on the K-nearest neighbor algorithm. Taking the LHK core rockfilldam in Southwest China as an example, we performed real-time monitoring for the spreading thickness and conducted real-time interactive queries regarding the spreading thickness. This approach provides a new method for controlling the spreading thickness of the core rockfilldam storehouse surface.
文摘Short-term traffic flow is one of the core technologies to realize traffic flow guidance. In this article, in view of the characteristics that the traffic flow changes repeatedly, a short-term traffic flow forecasting method based on a three-layer K-nearest neighbor non-parametric regression algorithm is proposed. Specifically, two screening layers based on shape similarity were introduced in K-nearest neighbor non-parametric regression method, and the forecasting results were output using the weighted averaging on the reciprocal values of the shape similarity distances and the most-similar-point distance adjustment method. According to the experimental results, the proposed algorithm has improved the predictive ability of the traditional K-nearest neighbor non-parametric regression method, and greatly enhanced the accuracy and real-time performance of short-term traffic flow forecasting.
基金LG was funded by the German Research Foundation(DFG 320926971)through the project“Analysis of diversity effects on above-groundproductivity in forests:advancing the mechanistic understanding of spatiotemporal dynamics in canopy space filling using mobile laser scanning”。
文摘Background:Species-specific genotypic features,local neighbourhood interactions and resource supply strongly influence the tree stature and growth rate.In mixed-species forests,diversity-mediated biomass allocation has been suggested to be a fundamental mechanism underlying the positive biodiversity-productivity relationships.Empirical evidence,however,is rare about the impact of local neighbourhood diversity on tree characteristics analysed at a very high level of detail.To address this issue we analysed these effects on the individual-tree crown architecture and tree productivity in a mature mixed forest in northern Germany.Methods:Our analysis considers multiple target tree species across a local neighbourhood species richness gradient ranging from 1 to 4.We applied terrestrial laser scanning to quantify a large number of individual mature trees(N=920)at very high accuracy.We evaluated two different neighbour inclusion approaches by analysing both a fixed radius selection procedure and a selection based on overlapping crowns.Results and conclusions:We show that local neighbourhood species diversity significantly increases crown dimension and wood volume of target trees.Moreover,we found a size-dependency of diversity effects on tree productivity(basal area and wood volume increment)with positive effects for large-sized trees(diameter at breast height(DBH)>40 cm)and negative effects for small-sized(DBH<40 cm)trees.In our analysis,the neighbour inclusion approach has a significant impact on the outcome.For scientific studies and the validation of growth models we recommend a neighbour selection by overlapping crowns,because this seems to be the relevant scale at which local neighbourhood interactions occur.Because local neighbourhood diversity promotes individual-tree productivity in mature European mixed-species forests,we conclude that a small-scale species mixture should be considered in management plans.
文摘As climate change negotiations progress,monitoring biomass and carbon stocks is becoming an important part of the current forest research.Therefore,national governments are interested in developing forest-monitoring strategies using geospatial technology.Among statistical methods for mapping biomass,there is a nonparametric approach called k-nearest neighbor(kNN).We compared four variations of distance metrics of the kNN for the spatially-explicit estimation of aboveground biomass in a portion of the Mexican north border of the intertropical zone.Satellite derived,climatic,and topographic predictor variables were combined with the Mexican National Forest Inventory(NFI)data to accomplish the purpose.Performance of distance metrics applied into the kNN algorithm was evaluated using a cross validation leave-one-out technique.The results indicate that the Most Similar Neighbor(MSN)approach maximizes the correlation between predictor and response variables(r=0.9).Our results are in agreement with those reported in the literature.These findings confirm the predictive potential of the MSN approach for mapping forest variables at pixel level under the policy of Reducing Emission from Deforestation and Forest Degradation(REDD+).
基金supported by the Social Science Foundation of China under Grant No.17BGL231。
文摘On the basis of machine leaning,suitable algorithms can make advanced time series analysis.This paper proposes a complex k-nearest neighbor(KNN)model for predicting financial time series.This model uses a complex feature extraction process integrating a forward rolling empirical mode decomposition(EMD)for financial time series signal analysis and principal component analysis(PCA)for the dimension reduction.The information-rich features are extracted then input to a weighted KNN classifier where the features are weighted with PCA loading.Finally,prediction is generated via regression on the selected nearest neighbors.The structure of the model as a whole is original.The test results on real historical data sets confirm the effectiveness of the models for predicting the Chinese stock index,an individual stock,and the EUR/USD exchange rate.
文摘The natural neighbour method can be considered as one of many variants of the meshless methods. In the present paper, a new approach based on the Fraeijs de Veubeke （FdV） functional, which is initially developed for linear elasticity, is extended to the case of geometrically linear but materially non-linear solids. The new approach provides an original treatment to two classical problems： the numerical evaluation of the integrals over the domain A and the enforcement of boundary conditions of the type ui = hi on Su. In the absence of body forces （Fi = 0）, it will be shown that the calculation of integrals of the type fA .dA can be avoided and that boundary conditions of the type ui = hi on Su can be imposed in the average sense in general and exactly if hi is linear between two contour nodes, which is obviously the case for tTi = O.
基金supported by the National Natural Science Foundation of China(Grant No.12002246 and No.52178301)Knowledge Innovation Program of Wuhan(Grant No.2022010801020357)+2 种基金the Science Research Foundation of Wuhan Institute of Technology(Grant No.K2021030)2020 annual Open Fund of Failure Mechanics&Engineering Disaster Prevention and Mitigation,Key Laboratory of Sichuan Province(Sichuan University)(Grant No.2020JDS0022)Open Research Fund Program of Hubei Provincial Key Laboratory of Chemical Equipment Intensification and Intrinsic Safety(Grant No.2019KA03)。
文摘This paper proposes an active learning accelerated Monte-Carlo simulation method based on the modified K-nearest neighbors algorithm.The core idea of the proposed method is to judge whether or not the output of a random input point can be postulated through a classifier implemented through the modified K-nearest neighbors algorithm.Compared to other active learning methods resorting to experimental designs,the proposed method is characterized by employing Monte-Carlo simulation for sampling inputs and saving a large portion of the actual evaluations of outputs through an accurate classification,which is applicable for most structural reliability estimation problems.Moreover,the validity,efficiency,and accuracy of the proposed method are demonstrated numerically.In addition,the optimal value of K that maximizes the computational efficiency is studied.Finally,the proposed method is applied to the reliability estimation of the carbon fiber reinforced silicon carbide composite specimens subjected to random displacements,which further validates its practicability.
文摘In the contemporary era of unprecedented innovations such as Internet of Things(IoT),modern applications cannot be imagined without the presence of Wireless Sensor Network(WSN).Nodes in WSN use neighbour discovery(ND)protocols to have necessary communication among the nodes.Neighbour discovery process is crucial as it is to be done with energy efficiency and minimize discovery latency and maximize percentage of neighbours discovered.The current ND approaches that are indirect in nature are categorized into methods of removal of active slots from wake-up schedules and intelligent addition of new slots.The two methods are found to have certain drawbacks.Thefirst category disturbs original integrity of wake-up schedules leading to reduced chances of discovering new nodes in WSN as neighbours.When second category is followed,it may have inefficient slots in the wake-up schedules leading to performance degradation.Therefore,the motivation behind the work in this paper is that by combining the two categories,it is possible to reap benefits of both and get rid of the limitations of the both.Making a hybrid is achieved by introducing virtual nodes that help maximize performance by ensuring original integrity of wake-up schedules and adding of efficient active slots.Thus a Hybrid Approach to Neighbour Discovery(HAND)protocol is realized in WSN.The simulation study revealed that HAND outperforms the existing indirect ND models.
基金Supported by the National Key Research and Development Program (No.2019YFA0707201)the Key Work Program of Institute of Scientific and Technical Information of China (No.ZD2022-01,ZD2023-07)。
文摘Stock trend prediction is a challenging problem because it involves many variables.Aiming at the problem that some existing machine learning techniques, such as random forest(RF), probabilistic random forest(PRF), k-nearest neighbor(KNN), and fuzzy KNN(FKNN), have difficulty in accurately predicting the stock trend(uptrend or downtrend) for a given date, a generalized Heronian mean(GHM) based FKNN predictor named GHM-FKNN was proposed.GHM-FKNN combines GHM aggregation function with the ideas of the classical FKNN approach.After evaluation, the comparison results elucidated that GHM-FKNN outperformed the other best existing methods RF, PRF, KNN and FKNN on independent test datasets corresponding to three stocks, namely AAPL, AMZN and NFLX.Compared with RF, PRF, KNN and FKNN, GHM-FKNN achieved the best performance with accuracy of 62.37% for AAPL, 58.25% for AMZN, and 64.10% for NFLX.
