Improved customer choice predictions using ensemble methods

In this paper various ensemble learning methods from machine learning and statistics are considered and applied to the customer choice modeling problem. The application of ensemble learning usually improves the prediction quality of flexible models like decision trees and thus leads to improved predictions. We give experimental results for two real-life marketing datasets using decision trees, ensemble versions of decision trees and the logistic regression model, which is a standard approach for this problem. The ensemble models are found to improve upon individual decision trees and outperform logistic regression.     

草酰胺分子构型、稳定性、IR光谱及分子轨道的密度泛函与从头算研究

运用多种密度泛函理论 (DFT)和从头算 (abinitio)方法研究了草酰胺顺、反异构体的分子几何构型 ,结果表明Beck88交换函数与LYP非局域相关函数形成杂化的DFT方法B3LYP与MP2方法计算结果比较吻合。在此基础上选择B3LYP方法对草酰胺分子电子结构、红外光谱和前线分子轨道组成进行了系统研究 ;与实验事实对比分析了气态草酰胺分子中配位原子的反应活性     

Investigation and reduction of variance of a weighted estimate in numerical statistical simulation

The efficiency of the “value” and “partial value” modeling related to the construction of a modeling distribution for an auxiliary random variable by multiplying the initial density by a value function is investigated. The value function usually corresponds to a solution of the adjoint equation. Conditions under which the value modeling of the initial distribution reduces the variance compared to the direct simulation are obtained. It is proved that the variance of the weighted estimate is bounded in the case of the partial value modeling. This proposition provides a basis for a method for determining whether or not the variance of the weighted estimate is bounded. This method uses the majorizing adjoint equation. Using a practically important problem in transport theory as an example, the asymptotic optimization of the distribution of the mean free path is presented. The application of the proposed method of the investigation of the variance boundedness for the analysis of the classical exponential transformation method of simulating the mean free path of a particle in the one-dimensional and the spherical variants is discussed.     

重力场与静电场特点的比较

“场”是一种客观存在的物质，但由于“场”具有不易直接感受的特殊存在形态，同学们对“场”的物质性的理解不够深刻，解决“场”的相关问题时也感到棘手．笔者就高中阶段最典型的两种场——“重力场”和“静电场”的力学性质与能量特点进行比较，使同学们对“场”的概念有更深刻的认识．     

杨氏模量的不确定度评定

本文介绍杨氏模量的不确定度计算方法。     

用方差定义的激光束横向尺寸的测量

用方差定义的激光束横向尺寸,有着普适、简单和便于利用光学元件并行运算功能的特点.本文提出一种借助改变柱形容器内吸收液浓度来直接测量各种波长激光束横向尺寸的理论和方法.     

Problems of an additional experiment

In some situations estimates of unknown parameters must be corrected by additional measurements. It is in principle no problem to calculate the corrected estimates, however, it is of more interest to find formulae for correction itself. The formulae enable us to design an additional experiment and to judge its usefulness. The aim of the paper is to find such formulae for several situations. Supported by the grant of the Council of Czech Government MSM 6 198 959 214.     

负关联噪声驱动下单模激光的定态分析

研究了负关联的加法和乘法高斯白噪声驱动下单模激光损失模型的定态情况。文中推导了负关联情况下，定态时的激光场幅的几率分布，光强的平均值，光强的协方差以及光强的偏斜率。并和文献〔１，２〕中正关联时的定态分析作了比较，发现了有意义的新现象。     

The diagnonal multivariate natural exponential families and their classification

A natural exponential family (NEF)F in ? ⁿ ,n>1, is said to be diagonal if there existn functions,a ₁,...,a _n , on some intervals of ?, such that the covariance matrixV _F (m) ofF has diagonal (a ₁(m ₁),...,a _n (m _n )), for allm=(m ₁,...,m _n ) in the mean domain ofF. The familyF is also said to be irreducible if it is not the product of two independent NEFs in ? ^k and ? ^n-k , for somek=1,...,n?1. This paper shows that there are only six types of irreducible diagonal NEFs in ? ⁿ , that we call normal, Poisson, multinomial, negative multinomial, gamma, and hybrid. These types, with the exception of the latter two, correspond to distributions well established in the literature. This study is motivated by the following question: IfF is an NEF in ? ⁿ , under what conditions is its projectionp(F) in ? ^k , underp(x ₁,...,x _n )∶=(x ₁,...,x _k ),k=1,...,n?1, still an NEF in ? ^k ? The answer turns out to be rather predictable. It is the case if, and only if, the principalk×k submatrix ofV _F (m ₁,...,m _n ) does not depend on (m _k+1,...,m _n ).     

An Efficient One-Arrow-Two-Hawks Strategy Achieves High Efficiency and Stable Batch Variance for Benzodifuran-based Polymer Solar Cells

With the development of organic solar cells (OSCs), the high-performance and stable batch variance are becoming a new challenge for designing polymer donors. To obtain high photovoltaic performance, adopting polymers with high molecular weight as donors is an ordinary strategy. However, the high molecular weight need to subtly control the reaction time and state, inevitably caused batch-to-batch variations. Herein, a strategy of steric effect is applied to benzodifuran (BDF)-based polymer by introducing different positions of Cl atom, producing two polymers PBDFCl-1 and PBDFCl-2. The more twisted side chains conformation not only achieve the control of moderate molecular weight for PBDFCl-2, but also easily form molecular stacking through adopting BDF unit and maintain sufficient polymeric crystallinity. Due to the optimized stacking mode and good blend miscibility, PBDFCl-2-based device exhibitsa more elegant power conversion efficiency (PCE) of 17.00% compared to PBDFCl-1-based device. This is the highest efficiency record for BDF-based binary OSCs. Meanwhile, the PCE device variation of the different molecular weights for PBDFCl-2 is little, indicating the reduction of the batch variation. Therefore, smartly using steric effect of Cl atom in strong crystalline BDF unit can form efficient molecular stacking regulations and realize the coordination of high-performance and stable batch variance.