摘要:
Abstract: This paper chooses the relevant characteristics of the customers who buy insurance as the research object, and based on the methods of correlation analysis and logistic regression, predicts the renewal rate of the people who buy insurance. Firstly, the data are preprocessed, and the available variables are transformed into virtual variables for correlation analysis. In this paper, two different types of data, discrete and continuous, are virtualized. For discrete data, discrete intervals or attributes can be directly transformed into “items”. For continuous data, the idea of iteration dichotomy is introduced, which is based on support and confidence. The continuous attribute values are divided into optimal intervals, and a new item is created for each different attribute value pair to obtain the virtual variables of the continuous attribute. After quantifying the data obtained from association rules, the association analysis is carried out, and all other variables including “whether to renew insurance” variables in strong association rules are selected as the six main influencing factors: vehicle age, renewal year, insured age, new car purchase price, signing premium and insurance premium. The logistic regression model is used to obtain the relationship between these influencing factors and renewal rate. Then we predicted again to get renewal rate value. The fitting degree is 96.7%. Then Z statistics is constructed to locate the attributes of customers by statistical inference, so as to achieve accurate customer portraits.#@#@#摘要: 本文选取购买保险的客户的相关特性作为研究对象,基于关联分析以及逻辑回归等方法,对百姓在购买保险方面的续保率进行预测。首先对数据进行预处理,将可用变量转化为虚拟变量以做关联分析,本文分别对离散与连续两种不同类型的数据进行虚拟化处理,针对离散型数据可直接将离散化区间或属性直接转化为“项”,针对连续型数据,本文引入迭代二划分的思想,基于支持度与置信度对连续属性值进行最优区间划分,为每个不同的属性值创建一个新的项来得到连续型属性的虚拟变量。将量化关联规则后得到的数据,对其进行关联分析,选取强关联规则中包含“是否续保”变量的其他所有变量:车龄,续保年,被保险人年龄,新车购置价,签单保费,三者险保费6种因素作为续保的主要影响因子,利用Logistic回归模型得到这些影响因素与续保率之间的关系,再预测得到续保率,其拟合度为96.7%。而后构建Z统计量,借助统计推断可为客户定位其属性,以实现精准的客户画像。
摘要:
针对目前油菜虫害识别在背景、角度、姿态、光照等方面的鲁棒性问题,提出一种基于深度卷积神经网络的油菜虫害检测方法:首先在卷积神经网络和区域候选网络的基础上,构建油菜虫害检测模型,再在深度学习tensorflow框架上实现模型的检测,最后对比分析结果。油菜虫害检测模型利用VGG16网络提取油菜虫害图像的特征,区域候选网络生成油菜害虫的初步位置候选框,Fast R–CNN实现候选框的分类和定位。结果表明,该方法可实现对蚜虫、菜青虫(幼虫)、菜蝽、跳甲、猿叶甲5种油菜害虫的快速准确检测,平均准确率达94.12%,与RCNN、 Fast R–CNN、多特征融合方法、颜色特征提取方法相比,准确率分别提高了28%、23%、12%、2%。
关键词:
加州电力市场;自相关;互相关;Granger因果检验 California Electricity Market
摘要:
Abstract: In order to explore the formation mechanism of electricity price in competitive electricity market, in this paper, the clearing price and load of California electricity market are considered as study object to detect the causal relationship between them by using Granger. Firstly, the long-range auto-correlation of real-time electricity price and quantity of the years of 1999 to 2000 as well as the cross-correlation between them are tested. Then, the first-order difference of the charge quantity and its corresponding price data for 24 periods in 1999 and 2000 is made respectively. After that, we investigate the stability of the sequence by ADF test. And then the causality test between the charge quantity and the price sequence after the first-order difference is carried out by Granger test. The results show that there is a significant correlation between the quantity and electricity price, which is causal to each other. Finally, some suggestions are provided for the healthy operation of electricity market of China.#@#@#摘要: 为探索竞争性电力市场电价形成机制,本文以美国加利福尼亚州(以下简称加州)电力市场出清价格、电荷量为研究对象,利用格兰杰方法检测二者之间的因果关系。首先选取1999~2000年的实时电价与电荷量进行长程自相关性、互相关性检测。然后分别将1999年和2000年全年24个时段的电荷量及其对应电价的数据进行一阶差分,选用ADF检验法检测序列的平稳性,而后利用Granger检验法对一阶差分后的电荷量与电价序列进行因果关系检验。结果表明,电荷量与电价的具有显著的相关关系,且互为因果。最后为我国电力市场健康运行提供了建议。
摘要:
In this study, the complete mitochondrial genome (mitogenome) of Gibbovalva kobusi was at first sequenced by high-throughput sequencing. As a circular DNA molecule, the complete mitogenome is 15,717 bp in length (GeneBank accession number: MK956103) and consists of 13 protein-coding genes (PCGs), 22 transfer RNA (tRNA) genes, 2 ribosomal RNA (rRNA) genes, and an AT-rich region. The nucleotide composition is A (41.0%), C (11.6%), G (7.9%), and T (39.5%). Based on the sequences of complete mitogenome from 11 species as ingroups and 3 superfamily Tineoidea species as outgroups, the phylogenetic trees were constructed. The family Gracilariidae as a monophyletic clade is strongly supported by the bootstrap value of 100%.
作者机构:
[Wu, Guochun] Huaqiao Univ, Fujian Prov Univ Key Lab Computat Sci, Sch Math Sci, Quanzhou 362021, Fujian, Peoples R China.;[Tan, Zhong] Xiamen Univ, Sch Math Sci, Xiamen 361005, Fujian, Peoples R China.;[Tan, Zhong] Xiamen Univ, Fujian Prov Key Lab Math Modeling & Sci Comp, Xiamen 361005, Fujian, Peoples R China.;[Xu, Jiankai] Hunan Agr Univ, Coll Informat Sci & Technol, Changsha 410128, Hunan, Peoples R China.
通讯机构:
School of Mathematical Sciences and Fujian Provincial Key Laboratory on Mathematical Modeling and Scientific Computing, Xiamen University, Xiamen, China
摘要:
We study the heat flow of equation of H-surface with non-zero Dirichlet boundary in the present article. Introducing the “stable set”
$$\mathfrak{M}_2$$
and “unstable set”
$$\mathfrak{M}_1$$
, we show that there exists a unique global solution provided the initial data belong to
$$\mathfrak{M}_2$$
and the global solution converges to zero in H1 exponentially as time goes to infinity. Moreover, we also prove that the local regular solution must blow up at finite time provided the initial data belong to
$$\mathfrak{M}_1$$
.
摘要:
Abstract: This article analyzes the annul use of synthetic opioids at the state and county levels in Kentucky, Ohio, Pennsylvania, Virginia, and West Virginia during the year 2010-2017. After carrying out the visual analysis of the data, the general transfer trend of two target substances can be initially determined, and the results show that the use of target substances is more serious at cities on the border. Then, we select the use of heroin as the reference sequence and make oxycodone, fentanyl, hydrocodeine and buprenorphine represent the other synthetic opioids according to the results of correlation analysis. Using the Markov chain model, the use of the target substances in five studied states can be predicted, and the threshold level of the number of opioids in five states is given as reference for the governors. The results show that the US government needs to concentrate more on the control of fentanyl drugs, and the use of opioids in West Virginia is relatively optimal.#@#@#摘要: 本文选取美国肯塔基州、俄亥俄州、宾夕法尼亚州、弗吉尼亚州和西弗吉尼亚州从2010年到2017年的县内阿片类药物鉴定数与州内阿片类药物鉴定数作为研究对象,并对该数据进行可视化分析,初步确定阿片类药物的大致转移趋势,结果显示阿片类药物主要集中在州的边境。然后利用关联度分析,以主要的阿片类药物海洛因作为参考标准,选取了海洛因、羟考酮、芬太尼、氢可待因和丁丙诺啡5种阿片类药物来衡量阿片类药物总体的使用情况。运用马尔可夫链模型,对美国5个州内阿片类药物的使用情况进行预测,给出5种阿片类药物数量的阈值水平,供美国政府参考。结果显示,美国政府需要加大力度对芬太尼药物进行控制,在西弗吉尼亚州的阿片类药物的使用情况相对最佳。