基于云相似度的B2B供应商个性化推荐算法

为满足B2B电子商务平台下供应商选择个性化服务需求,引入云模型中云的相似性度量算法和基于资源相似度的数据权重,对基于项目的协同过滤推荐算法进行改进,并用于供应商推荐过程中。实验表明,改进后的算法可以在一定程度上解决数据稀疏性问题并及时反映用户兴趣变化,从而实现个性化推荐服务,帮助企业快速有效开发供应商伙伴关系,提高企业的生产效率和竞争力。

基于关键字和序列匹配的自动评分算法的研究

自动评分在在线考试系统中起着至关重要的作用。通过分析传统编程题在手动和自动判分过程中存在的问题,提出了一种自动评分的改进算法。新算法对关键字进行权重分析,并在预处理的基础上将序列匹配算法进行改进,使得自动评分算法既能提高效率又能降低机械匹配带来的误判。通过将该算法应用在《C语言》在线考试系统中,结果表明,该方法从时效性、准确性上都是较有效的自动评分解决方案。

适应用户兴趣变化的协同过滤推荐算法

协同过滤算法是至今为止最成功的个性化推荐技术之一,被应用到很多领域中.但传统协同过滤算法不能及时反映用户的兴趣变化.针对这个问题,提出两种改进度量:基于时间的数据权重和基于资源相似度的数据权重,在此基础上将它们有机结合,并将这两种权重引入基于资源的协同过滤算法的生成推荐过程中.实验表明,改进后的算法比传统协同过滤算法在推荐准确度上有明显提高.

APPROACH FOR FUZZY MULTI-ATTRIBUTE DECISION-MAKING WITH FUZZY COMPLEMENTARY PREFERENCE RELATION ON ALTERNATIVES

In presented fuzzy multi-attribute decision-making （FMADM） problems, the information about attribute weights is interval numbers and the decision maker （DM） has fuzzy complementary preference relation on alternatives. Firstly, the decision-making information based on the subjective preference information in the form of the fuzzy complementary judgment matrix is uniform by using a translation function. Then an objective programming model is established. Attribute weights are obtained by solving the model, thus the fuzzy overall values of alternatives are derived by using the additive weighting method. Secondly, the ranking approach of alternatives is proposed based on the degree of similarity between the fuzzy positive ideal solution of alternatives （FPISA） and the fuzzy overall values. The method can sufficiently utilize the objective information of alternatives and meet the subjective requirements of the DM as much as possible. It is easy to be operated and implemented on a computer. Finally, the proposed method is applied to the project evaluation in the venture investment.

反映用户兴趣变化的协同过滤算法

针对传统协同过滤算法存在的两个弊端:一是传统的相似性度量方法在评分矩阵稀疏的情况下很难准确地反映用户间的相似性,二是不能及时反映用户的兴趣变化,提出一种新的相似性计算方法。此方法把基于用户兴趣度的相似性度量与基于项目相似度的数据权重结合,形成一种考虑用户兴趣变化的相似性度量方法。实验结果表明,改进后的算法集成了上述两种方法的优点,对传统算法中存在的两个弊端进行了改善,在推荐准确度上有所提高。

Improved random noise attenuation using f-x empirical mode decomposition and local similarity

Conventional f-x empirical mode decomposition（EMD） is an effective random noise attenuation method for use with seismic profiles mainly containing horizontal events.However,when a seismic event is not horizontal,the use of f-x EMD is harmful to most useful signals.Based on the framework of f-x EMD,this study proposes an improved denoising approach that retrieves lost useful signals by detecting effective signal points in a noise section using local similarity and then designing a weighting operator for retrieving signals.Compared with conventional f-x EMD,f-x predictive filtering,and f-x empirical mode decomposition predictive filtering,the new approach can preserve more useful signals and obtain a relatively cleaner denoised image.Synthetic and field data examples are shown as test performances of the proposed approach,thereby verifying the effectiveness of this method.

Risk assessment model of tunnel water inrush based on improved attribute mathematical theory

Tunnel water inrush is one of the common geological disasters in the underground engineering construction.In order to effectively evaluate and control the occurrence of water inrush,the risk assessment model of tunnel water inrush was proposed based on improved attribute mathematical theory.The trigonometric functions were adopted to optimize the attribute mathematical theory,avoiding the influence of mutation points and linear variation zones in traditional linear measurement functions on the accuracy of the model.Based on comprehensive analysis of various factors,five parameters were selected as the evaluation indicators for the model,including tunnel head pressure,permeability coefficient of surrounding rock,crushing degree of surrounding rock,relative angle of joint plane and tunnel section size,under the principle of dimension rationality,independence,directness and quantification.The indicator classifications were determined.The links among measured data were analyzed in detail,and the objective weight of each indicator was determined by using similar weight method.Thereby the tunnel water inrush risk assessment model is established and applied in four target segments of two different tunnels in engineering.The evaluation results and the actual excavation data agree well,which indicates that the model is of high credibility and feasibility.

A feature selection approach based on a similarity measure for software defect prediction

Software defect prediction is aimed to find potential defects based on historical data and software features. Software features can reflect the characteristics of software modules. However, some of these features may be more relevant to the class （defective or non-defective）, but others may be redundant or irrelevant. To fully measure the correlation between different features and the class, we present a feature selection approach based on a similarity measure （SM） for software defect prediction. First, the feature weights are updated according to the similarity of samples in different classes. Second, a feature ranking list is generated by sorting the feature weights in descending order, and all feature subsets are selected from the feature ranking list in sequence. Finally, all feature subsets are evaluated on a k-nearest neighbor （KNN） model and measured by an area under curve （AUC） metric for classification performance. The experiments are conducted on 11 National Aeronautics and Space Administration （NASA） datasets, and the results show that our approach performs better than or is comparable to the compared feature selection approaches in terms of classification performance.