Principal weighted logistic regression for sufficient dimension reduction in binary classification

Sufficient dimension reduction (SDR) is a popular supervised machine learning technique that reduces the predictor dimension and facilitates subsequent data analysis in practice. In this article, we propose principal weighted logistic regression (PWLR), an efficient SDR method in binary classification where inverse-regression-based SDR methods often suffer. We first develop linear PWLR for linear SDR and study its asymptotic properties. We then extend it to nonlinear SDR and propose the kernel PWLR. Evaluations with both simulated and real data show the promising performance of the PWLR for SDR in binary classification.     

“有效行”特征对手写体字符的识别

从无限制的手写体数字的结构出发，提出“有效行”特征的概念及其提取算法。该特征具有维数小、平移不变、字符小角度旋转不变等特点。建立了相应的字符特征库，利用ＢＰ网络对样本字符进行研究，通过大量的手写体数字识别测试表明：该方法在识别速度上优于利用单一的矩特征、小波特征等传统识别方法；在误识率方面也优于一般的单一识别方法。     

人脸识别中多目标最优不相关图像鉴别分析研究

考虑图像投影鉴别分析问题,为提高特征抽取的速度和识别率,利用图像矩阵直接构造图像散布矩阵,在具有统计不相关的条件下将Foley-Sammon鉴别分析(FSLDA)转化为两目标约束优化问题,并给出了有效投影向量的概念;根据多目标优化的最优性条件可将求取有效投影向量的问题归结为求广义特征方程的最大特征值对应的特征向量,并据此进行特征抽取,进而提出了两目标最优图像投影鉴别分析方法。与其他鉴别投影分析方法相比,该方法具有以下特点:(1)可直接由图像矩阵构建散布矩阵;(2)有效投影向量具有统计不相关性;(3)训练样本的类内散布矩阵不必为可逆的,也不需要求某种形式矩阵的逆。在ORL标准人脸库和NUST603人脸库上的试验结果表明,上述图像投影鉴别分析方法在识别性能上较以往的方法有一定的提高,尤其是特征抽取的速度有明显的提高。     

蒲公英多糖提取工艺的研究

通过对蒲公英多糖提取的单因素试验和正交实验研究,得出蒲公英多糖的热水浸提最佳的工艺条件.     

Mining the mind: emotional extraction,productivity, and predictability in the twenty-first century

Surveying recent developments in management and work culture, computing and social media, and science and psychology, this article speculates on the concept of emotional extraction. Emotional extraction is defined in two ways. One iteration involves the transfer of emotional resources from one individual or group to another, such as that which occurs in the work of caring for others, but which also increasingly occurs in the work of producing new technology, such as emotionally aware computers. A second instance of emotional extraction entails the use of emotion knowledge – or theories about emotions, such as emotional intelligence – to generate conclusions or predictions about human behaviour. Emotional extraction in service work, management, marketing, social media, artificial intelligence, and neuroscience are discussed. ‘Mining the mind’ focuses in particular on emotional extraction that enhances both productivity and predictability, in turn tracing how emotionally extractive sites are implicated within the production and hierarchical valuation of difference – especially racial and gendered, but also neural difference – in everyday life. The article aims to offer scholars in cultural studies, as well as critical race theory, feminist theory, and critical disability studies, ways to think about this newly intensifying resource extraction and the intersections of culture, capital, and human experience that such extraction indexes and makes possible.     

当代刑讯逼供之刑事政策及启示

刑讯逼供在中国属于非法行为,但由于＂活法＂的力量使其在现实层面仍顽强存在。尽管当代刑事政策对刑讯逼供的禁止具有某种不彻底性,但仍然体现出巨大的历史进步意义。促使刑讯逼供刑事政策变动的真正原因并非启蒙时代人权教义或原则的提出,以科学技术为代表的人类物质文明的迅猛发展才使人道观念的实现具有可能性。只有科技进步才是遏制刑讯逼供的决定性力量。     

A Benchmarking Approach to Forecast Combination

This article is concerned with the development of a statistical model-based approach to optimally combine forecasts derived from an extrapolative model, such as an autoregressive integrated moving average (ARIMA) time series model, with forecasts of a particular characteristic of the same series obtained from independent sources. The methods derived combine the strengths of all forecasting approaches considered in the combination scheme. The implications of the general theory are investigated in the context of some commonly encountered seasonal ARIMA models. An empirical example to illustrate the method is included.     

蓝靛果总黄酮的超声波提取研究

采用超声波辅助法提取蓝靛果总黄酮.采用单因素试验探讨提取温度、提取时间、料液比、乙醇溶液体积分数、提取次数对蓝靛果总黄酮提取量的影响,采用正交试验优化提取工艺.结果表明:蓝靛果总黄酮的最佳提取工艺条件为提取温度75℃、提取时间30 min、料液比1∶15 (g/mL)、乙醇溶液体积分数65％、超声波功率100W、提取次数1,此条件下总黄酮提取量为15.14mg/g.     

二维小波变换在虹膜特征提取中的应用研究

本文提出了一种采用canny变换来提取虹膜特征,并用小波神经网络来进行分类的虹膜检测识别的新方法。该方法结合了小波神经网络作为一种新的分类方法,它很好地改进了识别精度,同时提高了系统的性能。一个简洁并且能快速训练的算子Adaboost也将在小波神经网络中给出介绍。实验结果表明,应用该算子进行分类识别时,识别率为100%,该方法具有很高效的可行性。     

蒲公英多糖提取工艺的研究

通过对蒲公英多糖提取的单因素试验和正交实验研究，得出蒲公英多糖的热水浸提最佳的工艺条件。