近红外光谱结合共识模型快速检测果酒的总酚含量

果酒发酵中的多酚是引起果酒口感、颜色变化的重要因素。为保证果酒品质,有必要开发一种快速监测发酵过程中多酚含量变化的技术。收集不同批次成熟期的蓝莓、桑葚为原料,分别碾压成汁,同时按比例混合二者,于小型发酵罐进行发酵。通过离线收集不同发酵时段的发酵液于离心管,高速离心后取上清液置于棕色瓶保存,共计得到48个果酒发酵样本。将上清液置于三个平行样比色皿,以傅里叶快速变换近红外光谱仪(FT-NIR)采集其透射光谱,取平均值作为该样本的光谱信号。然后将棕色瓶内的发酵液以国标法(即以标准液的吸光度值制定标准曲线)测定各样品的总酚含量,以duplex法计算样本光谱之间的距离且按2∶1的比例划分为训练集和预测集。采用间隔偏最小二乘法(iPLS)将训练集样本的透射光谱与总酚含量之间构建定量模型,间隔数从2依次变化到60个。该研究创新之处是使用共识方法融合多个已构建好的iPLS成员模型,按一定的共识规则分配权系数。通过各成员模型交互验证的残差及其残差之间的相关性来优化各成员模型的线性组合,以拉格朗日乘数法求解各成员模型的权系数,使间隔偏最小二乘-共识模型(consensual iPLS,CiPLS)的交互验证均方根误差最小。相比于全局PLS模型、划分不同间隔数量时的iPLS模型,CiPLS均具有较小的预测误差。当划分39个间隔时由三个iPLS成员模型(即14th,16th,18th)组成的共识模型误差最小为124.2,交互验证相关系数为0.944,对预测集样本的预测均方根误差为163.4,预测相关系数为0.931,预测性能均优于PLS和iPLS模型。另外,作为对比选用连续投影算法与无信息变量剔除法来优化光谱模型,其预测性能均不及本文提出的共识模型。分析各iPLS模型预测残差之间的相关性,发现共识模型主要是融合那些具有较高预测性能且模型间较低相关性的成员模型。结果表明,光谱分析结合共识方法可提高回归模型的预测精度、减少建模所需变量数,能够用于果酒总酚含量的离线快速检测。     

A consensus successive projections algorithm – multiple linear regression method for analyzing near infrared spectra

The successive projections algorithm (SPA) is widely used to select variables for multiple linear regression (MLR) modeling. However, SPA used only once may not obtain all the useful information of the full spectra, because the number of selected variables cannot exceed the number of calibration samples in the SPA algorithm. Therefore, the SPA-MLR method risks the loss of useful information. To make a full use of the useful information in the spectra, a new method named “consensus SPA-MLR” (C-SPA-MLR) is proposed herein. This method is the combination of consensus strategy and SPA-MLR method. In the C-SPA-MLR method, SPA-MLR is used to construct member models with different subsets of variables, which are selected from the remaining variables iteratively. A consensus prediction is obtained by combining the predictions of the member models. The proposed method is evaluated by analyzing the near infrared (NIR) spectra of corn and diesel. The results of C-SPA-MLR method showed a better prediction performance compared with the SPA-MLR and full-spectra PLS methods. Moreover, these results could serve as a reference for combination the consensus strategy and other variable selection methods when analyzing NIR spectra and other spectroscopic techniques.     

A model validation and consensus building environment

Over half of the failures in drug development are due to problems with the absorption, distribution, metabolism, excretion, and toxicity, or ADME/Tox properties of a candidate compound. The utilization of in silico tools to predict ADME/Tox and physicochemical properties holds great potential for reducing the attrition rate in drug research and development, as this technology can prioritize candidate compounds in the pharmaceutical R&D pipeline. However, a major concern surrounding the use of in silico ADME/Tox technology is the reliability of the property predictions. Bio-Rad Laboratories, Inc. has created a computational environment that addresses these concerns. This environment is referred to as KnowItAll®. Within this platform are encoded a number of ADME/Tox predictors, the ability to validate these predictors with/without in-house data and models, as well as build a ‘consensus’ model that may be a much better model than any of the individual predictive model. The KnowItAll® system can handle two types of predictions: real number and categorical classification.     

Chaos in social learning with multiple true states

Most existing social learning models assume that there is only one underlying true state. In this work, we consider a social learning model with multiple true states, in which agents in different groups receive different signal sequences generated by their corresponding underlying true states. Each agent updates his belief by combining his rational self-adjustment based on the external signals he received and the influence of his neighbors according to their communication. We observe chaotic oscillation in the belief evolution, which implies that neither true state could be learnt correctly by calculating the largest Lyapunov exponents and Hurst exponents.     

Amalgamation of partitions from multiple segmentation bases: A comparison of non-model-based and model-based methods

The segmentation of customers on multiple bases is a pervasive problem in marketing research. For example, segmentation service providers partition customers using a variety of demographic and psychographic characteristics, as well as an array of consumption attributes such as brand loyalty, switching behavior, and product/service satisfaction. Unfortunately, the partitions obtained from multiple bases are often not in good agreement with one another, making effective segmentation a difficult managerial task. Therefore, the construction of segments using multiple independent bases often results in a need to establish a partition that represents an amalgamation or consensus of the individual partitions. In this paper, we compare three methods for finding a consensus partition. The first two methods are deterministic, do not use a statistical model in the development of the consensus partition, and are representative of methods used in commercial settings, whereas the third method is based on finite mixture modeling. In a large-scale simulation experiment the finite mixture model yielded better average recovery of holdout (validation) partitions than its non-model-based competitors. This result calls for important changes in the current practice of segmentation service providers that group customers for a variety of managerial goals related to the design and marketing of products and services.     

不同预处理方法对苯丙酮尿症FTIR/ATR光谱筛查模型的影响

建立苯丙酮尿症的FTIR/ATR光谱筛查模型,比较基线校正、平滑、求导、傅里叶退卷积等光谱预处理方法对模型精度的影响。利用多模型共识偏最小二乘法(cPLS)建立干血片中苯丙氨酸浓度的校正模型,以相关系数(r)、预测均方根误差(RMSEP)、平均相对误差(MRE)和预测准确率(Acc)等指标,考察不同预处理方法对建模效果的影响。结果 一阶微分9点平滑处理方法效果最好。与原始光谱相比,模型的r,RMSEP,MRE和Acc分别从0.822 7,115.8,0.395和94.6改善到0.889 9,102.2,0.286和100。本方法直接快速、 不消耗试剂、 不产生污染,有望成为PKU大人群快速筛查的简便、 绿色新技术。     

Wilson's theorem for consensus functions on hierarchies

Wilson in 1972 gave a generalization of Arrow's theorem without the Pareto axiom. We prove a version of Wilson's theorem for consensus functions on hierarchies.     

Dynamic behavior of discrete-time multiagent systems with general communication structures

In this paper, we discuss the dynamic behavior of networks of dynamic agents with general communication topologies. We first analyze the basic case: systems with communication topologies that have spanning trees, i.e., the systems that solve consensus problems. We establish an algebraic condition to characterize each agent's contributions to the final state. And we also study the influence of time-delays on each agent's contributions. Then, we investigate the general case: systems with weakly connected topologies. By using matrix theory, we prove that the states of internal agents will converge to a convex combination of boundary agents in the absence or presence of communication time-delays, and we also show that the coefficients of the convex combination are independent of time-delays even if the delays are time-varying. These results have broad applications in other areas, e.g., study of swarm behavior, formation control of vehicles, etc.     

基于合意空间的模糊子群

利用合意空间(ConsensusSpace)理论给出了一个群的C模糊子群的定义.指出这种模糊子群实际上是基于t范(Tm范)的模糊子群.证明了Rosenfeld的模糊子群是C模糊子群,且每一个C模糊子群都与一类特殊的C模糊子群同构.从而为模糊子群提供了新的理论基础.     

Consensus problems in discrete-time multiagent systems with fixed topology

In this paper, we study the consensus problems in discrete-time multiagent systems with fixed topology. A necessary and sufficient condition for a system that solves a consensus problem is established, and the structure of consensus functions is characterized. Based on them, we introduce the standard topologies (graphs) of information flow, with which the systems can be viewed as single-leader-multi-follower systems. Moreover, the convex combination of these topologies can create a system that solves any predeterminate consensus problem. Additionally, we characterize the structural decomposition—the leaders-followers decomposition of a multiagent system, and establish a necessary and sufficient condition for an agent to be a leader.