The exponential pace of the spread of the digital world has served as one of the assisting forces to generate an enormous amount of informationflow-ing over the network.The data will always remain under the threat of t...The exponential pace of the spread of the digital world has served as one of the assisting forces to generate an enormous amount of informationflow-ing over the network.The data will always remain under the threat of technolo-gical suffering where intruders and hackers consistently try to breach the security systems by gaining personal information insights.In this paper,the authors pro-posed the HDTbNB(Hybrid Decision Tree-based Naïve Bayes)algorithm tofind the essential features without data scaling to maximize the model’s performance by reducing the false alarm rate and training period to reduce zero frequency with enhanced accuracy of IDS(Intrusion Detection System)and to further analyze the performance execution of distinct machine learning algorithms as Naïve Bayes,Decision Tree,K-Nearest Neighbors and Logistic Regression over KDD 99 data-set.The performance of algorithm is evaluated by making a comparative analysis of computed parameters as accuracy,macro average,and weighted average.Thefindings were concluded as a percentage increase in accuracy,precision,sensitiv-ity,specificity,and a decrease in misclassification as 9.3%,6.4%,12.5%,5.2%and 81%.展开更多
As extensions of means, expectiles embrace all the distribution information of a random variable.The expectile regression is computationally friendlier because the asymmetric least square loss function is differentiab...As extensions of means, expectiles embrace all the distribution information of a random variable.The expectile regression is computationally friendlier because the asymmetric least square loss function is differentiable everywhere. This regression also enables effective estimation of the expectiles of a response variable when potential explanatory variables are given. In this study, we propose the partial functional linear expectile regression model. The slope function and constant coefficients are estimated by using the functional principal component basis. The convergence rate of the slope function and the asymptotic normality of the parameter vector are established. To inspect the effect of the parametric component on the response variable, we develop Wald-type and expectile rank score tests and establish their asymptotic properties. The finite performance of the proposed estimators and test statistics are evaluated through simulation study. Results indicate that the proposed estimators are comparable to competing estimation methods and the newly proposed expectile rank score test is useful. The methodologies are illustrated by using two real data examples.展开更多
In this paper,we study the large-scale inference for a linear expectile regression model.To mitigate the computational challenges in the classical asymmetric least squares(ALS)estimation under massive data,we propose ...In this paper,we study the large-scale inference for a linear expectile regression model.To mitigate the computational challenges in the classical asymmetric least squares(ALS)estimation under massive data,we propose a communication-efficient divide and conquer algorithm to combine the information from sub-machines through confidence distributions.The resulting pooled estimator has a closed-form expression,and its consistency and asymptotic normality are established under mild conditions.Moreover,we derive the Bahadur representation of the ALS estimator,which serves as an important tool to study the relationship between the number of submachines K and the sample size.Numerical studies including both synthetic and real data examples are presented to illustrate the finite-sample performance of our method and support the theoretical results.展开更多
Photon mapping is widely used for global illumi- nation rendering because of its high computational efficiency. But its efficiency is still limited, mainly by the intensive sam- piing required in final gathering, a pr...Photon mapping is widely used for global illumi- nation rendering because of its high computational efficiency. But its efficiency is still limited, mainly by the intensive sam- piing required in final gathering, a process that is critical for removing low frequency artifacts of density estimation. In this paper, we propose a method to predict the final gather- ing estimation with direct density estimation, thereby achiev- ing high quality global illumination by photon mapping with high efficiency. We first sample the irradiance of a subset of shading points by both final gathering and direct radiance es- timation. Then we use the samples as a training set to predict the final gathered irradiance of other shading points through regression. Consequently, we are able to achieve about three times overall speedup compared with straightforward final gathering in global illumination computation with the same rendering quality.展开更多
文摘The exponential pace of the spread of the digital world has served as one of the assisting forces to generate an enormous amount of informationflow-ing over the network.The data will always remain under the threat of technolo-gical suffering where intruders and hackers consistently try to breach the security systems by gaining personal information insights.In this paper,the authors pro-posed the HDTbNB(Hybrid Decision Tree-based Naïve Bayes)algorithm tofind the essential features without data scaling to maximize the model’s performance by reducing the false alarm rate and training period to reduce zero frequency with enhanced accuracy of IDS(Intrusion Detection System)and to further analyze the performance execution of distinct machine learning algorithms as Naïve Bayes,Decision Tree,K-Nearest Neighbors and Logistic Regression over KDD 99 data-set.The performance of algorithm is evaluated by making a comparative analysis of computed parameters as accuracy,macro average,and weighted average.Thefindings were concluded as a percentage increase in accuracy,precision,sensitiv-ity,specificity,and a decrease in misclassification as 9.3%,6.4%,12.5%,5.2%and 81%.
基金supported by National Natural Science Foundation of China(Grant No.11771032)Natural Science Foundation of Shanxi Province of China(Grant No.201901D111279)+1 种基金the Research Grant Council of the Hong Kong Special Administration Region(Grant Nos.14301918 and 14302519)。
文摘As extensions of means, expectiles embrace all the distribution information of a random variable.The expectile regression is computationally friendlier because the asymmetric least square loss function is differentiable everywhere. This regression also enables effective estimation of the expectiles of a response variable when potential explanatory variables are given. In this study, we propose the partial functional linear expectile regression model. The slope function and constant coefficients are estimated by using the functional principal component basis. The convergence rate of the slope function and the asymptotic normality of the parameter vector are established. To inspect the effect of the parametric component on the response variable, we develop Wald-type and expectile rank score tests and establish their asymptotic properties. The finite performance of the proposed estimators and test statistics are evaluated through simulation study. Results indicate that the proposed estimators are comparable to competing estimation methods and the newly proposed expectile rank score test is useful. The methodologies are illustrated by using two real data examples.
文摘In this paper,we study the large-scale inference for a linear expectile regression model.To mitigate the computational challenges in the classical asymmetric least squares(ALS)estimation under massive data,we propose a communication-efficient divide and conquer algorithm to combine the information from sub-machines through confidence distributions.The resulting pooled estimator has a closed-form expression,and its consistency and asymptotic normality are established under mild conditions.Moreover,we derive the Bahadur representation of the ALS estimator,which serves as an important tool to study the relationship between the number of submachines K and the sample size.Numerical studies including both synthetic and real data examples are presented to illustrate the finite-sample performance of our method and support the theoretical results.
文摘Photon mapping is widely used for global illumi- nation rendering because of its high computational efficiency. But its efficiency is still limited, mainly by the intensive sam- piing required in final gathering, a process that is critical for removing low frequency artifacts of density estimation. In this paper, we propose a method to predict the final gather- ing estimation with direct density estimation, thereby achiev- ing high quality global illumination by photon mapping with high efficiency. We first sample the irradiance of a subset of shading points by both final gathering and direct radiance es- timation. Then we use the samples as a training set to predict the final gathered irradiance of other shading points through regression. Consequently, we are able to achieve about three times overall speedup compared with straightforward final gathering in global illumination computation with the same rendering quality.
