面向新闻领域的中文实体关系抽取

随着互联网技术的快速发展,人们能够及时地获取大量的新闻文本信息,如何从新闻中自动获取关键信息,把新闻中具有价值的信息转化为结构化数据,从而快速有效地获取有用的知识已是迫切需求。实体关系抽取是获取关键信息的方法之一,但目前关于中文的实体关系抽取工作较少。针对基于长短时记忆网络的中文实体识别模型难于提取长距离的依存关系特征和句法特征问题,提出利用双向树形长短时记忆神经网络提取依存句法树的结构特征。在提取的特征的基础上,使用条件随机场判断实体的类别和边界,并在实体识别模型中加入注意力机制提高模型的性能。在《人民日报》数据集和ACE 2005语料库上训练模型,验证了模型的有效性。     

The importance of atmospherics in the choice of hypermarkets and supermarkets

Are the atmospheric variables equally relevant for supermarket and hypermarket customers? As a contribution to the customers' choice of the store format, firstly this work aims to determine the perception of differences between hypermarkets and supermarkets in terms of store attributes. Second, it sheds light on the role played by atmospherics in the customers' decision process. This study uses a logit approach to model the probabilistic choice of the customers' store format using atmospheric and other marketing variables as important attributes of the grocery store. The model's estimation is based on personal interviews conducted during the shopping process in urban supermarkets and hypermarkets belonging to the same retail chain. The findings show that although the atmospheric variables are significantly important for both store formats, the customers perceive differently the importance of each retail atmospheric cue and other marketing variables, when buying in a hypermarket or in a supermarket: the choice of buying in hypermarket is explained by the significant importance given to the latent variables named pleasant atmosphere and accessibility and responsiveness; in contrast, the choice of buying in supermarkets is explained by the significant importance given to the latent variable nice decoration and empathy with the staff. The other latent variable designated by other marketing variables is only significantly important in the decision of choosing supermarkets.     

农业信贷对我国农业经济增长贡献的实证分析

文章利用面板数据基础上的随机效果模型对2000￣2003年间我国农业信贷对农业经济增长的贡献作出实证分析。研究结果表明,我国农业信贷与农业经济增长之间存在正相关关系,增加农业信贷投入,可以提高农业产出水平。因此应不断改善农业信贷环境,增加农业信贷供给,以促进农业产出的增加。     

基于随机森林算法的洪水灾害风险评估研究

根据流域灾害系统理论,在考虑致灾因子、孕灾环境和承灾体的基础上,选取9个风险评价指标,运用样本数据进行人工识别风险并得到训练样本,采用随机森林算法构建基于随机森林的洪水灾害风险评估模型。然后采用随机森林自评估工具,分析建立的洪水灾害风险评估模型的误差和指标,同时构建支持向量机模型作为对比方案,并采用五折交叉验证方法对基于随机森林算法的洪水灾害风险评估模型和支持向量机模型进行验证。最后以海河流域邱庄段为研究对象,分别运用基于随机森林算法的洪水灾害风险评估模型和基于支持向量机模型对相同的数据集进行评估和对比,结果显示,12 h内降雨总量、洪水持续时间和土壤含水量是引发洪水的主要因素,而基于随机森林算法的洪水灾害风险评估的训练精度及测试精度均高于支持向量机模型。     

A Conjoint/Logit Analysis of Consumers’ Responses to Genetically Modified Tofu in Taiwan

We use data collected from a consumer survey of face‐to‐face interviews to determine consumer demand for genetically modified (GM) tofu attributes in Taiwan. Conjoint analysis using logit models reveals that, on average, brand is the most important attribute in terms of influence on consumers’ preferences, followed by price, with GM content having the least influence. However, the analysis also reveals three distinct market segments for tofu. Apart from the largest segment (which reflects the average lack of concern about GM food), the remaining two segments are split according to their preference or antipathy towards GM tofu. This result suggests that GM labelling is helpful to Taiwanese tofu consumers.     

Willingness to Pay for Rural Landscape Improvements: Combining Mixed Logit and Random‐Effects Models

This paper reports the findings from a discrete‐choice experiment designed to estimate the economic benefits associated with rural landscape improvements in Ireland. Using a mixed logit model, the panel nature of the dataset is exploited to retrieve willingness‐to‐pay values for every individual in the sample. This departs from customary approaches in which the willingness‐to‐pay estimates are normally expressed as measures of central tendency of an a priori distribution. Random‐effects models for panel data are subsequently used to identify the determinants of the individual‐specific willingness‐to‐pay estimates. In comparison with the standard methods used to incorporate individual‐specific variables into the analysis of discrete‐choice experiments, the analytical approach outlined in this paper is shown to add considerable explanatory power to the welfare estimates.     

图像的薛定谔变换初探

介绍了图像的薛定谔变换的定义,并研究了图像的薛定谔变换性质在图像处理与分析的应用。在图像中多个目标的内部和外部区域,利用图像的薛定谔变换得到,实现多目标轮廓的自动提取,以及实现图像修描。最后用实验证明了该方法的有效性。     

构建基于VBA的PowerPoint随机指派系统

课堂提问是教学过程重要一环,通过师生问答互动,教师可以了解学生的接受程度,同时也鼓励学生积极思考。但现状是,教师经常面对学生遇见提问就低头的尴尬境地。针对在大学课堂教学过程中教师提问所面临的学生“集体沉默”这个普遍现象,文中利用VBA在PowerPoint中实现了一个随机指派系统,通过该系统随机指定回答问题的学生,提高了学生思考问题的积极性,活跃了课堂气氛。     

政府随机审计的动态博弈模型及政策启示

本文在一个三阶段动态博弈框架下,对政府随机审计过程中的政府审计机关与政府审计对象以及国家与政府审计对象之间的动态博弈特征进行了分析。分析结果显示：政府审计机关与政府审计对象之间存在一个纯策略纳什均衡,而国家和政府审计对象之间存在一个混合策略纳什均衡。政府审计对象接受国家实质性审计的概率是政府审计对象被政府机关审计发现违规或不作为后的惩罚倍数的单调递减函数,而政府审计对象的违规或不作为的概率分别是政府审计对象被政府机关审计发现违规或不作为后的惩罚倍数和政府审计对象的应尽义务或责任的单调递减函数,却是政府审计机关审计行为成本的单调递增函数。同时,结合研究结论和中国政府审计全覆盖的现实,给出了一些政策建议。     

基于效应的随机灰规划模型和应用

为了解决同时含有随机因素和灰色因素的不确定规划问题,通过结合区间灰数所属区间两个端点的随机性,给出随机区间灰数和随机区间灰函数的定义,提出了随机灰规划模型。通过综合效应函数理论用随机变量期望值和方差综合量化表示灰数所属区间的两个端点值。应用该理论对综合量化后的两个端点值继续进行综合量化,从而将随机灰规划转化为确定型规划问题。应用遗传算法进行求解。通过综合效应函数的理念,综合随机变量的期望和方差,同时综合区间灰数的区间因素,将随机灰规划数学模型转化为确定型规划模型即基于效应的随机灰规划模型。通过选取不同的综合效应函数,得到了关于不同决策意识下的随机灰规划的最优解。这个方法可为决策者进行不确定决策提供参考。