Feature-level sentiment analysis by using comparative domain corpora

Feature-level sentiment analysis (SA) is able to provide more fine-grained SA on certain opinion targets and has a wider range of applications on E-business. This study proposes an approach based on comparative domain corpora for feature-level SA. The proposed approach makes use of word associations for domain-specific feature extraction. First, we assign a similarity score for each candidate feature to denote its similarity extent to a domain. Then we identify domain features based on their similarity scores on different comparative domain corpora. After that, dependency grammar and a general sentiment lexicon are applied to extract and expand feature-oriented opinion words. Lastly, the semantic orientation of a domain-specific feature is determined based on the feature-oriented opinion lexicons. In evaluation, we compare the proposed method with several state-of-the-art methods (including unsupervised and semi-supervised) using a standard product review test collection. The experimental results demonstrate the effectiveness of using comparative domain corpora.     

Personalised news filtering and recommendation system using Chi-square statistics-based K-nearest neighbour (χ2SB-KNN) model

Recommendation problem has been extensively studied by researchers in the field of data mining, database and information retrieval. This study presents the design and realisation of an automated, personalised news recommendations system based on Chi-square statistics-based K-nearest neighbour (χ²SB-KNN) model. The proposed χ²SB-KNN model has the potential to overcome computational complexity and information overloading problems, reduces runtime and speeds up execution process through the use of critical value of χ² distribution. The proposed recommendation engine can alleviate scalability challenges through combined online pattern discovery and pattern matching for real-time recommendations. This work also showcases the development of a novel method of feature selection referred to as Data Discretisation-Based feature selection method. This is used for selecting the best features for the proposed χ²SB-KNN algorithm at the preprocessing stage of the classification procedures. The implementation of the proposed χ²SB-KNN model is achieved through the use of a developed in-house Java program on an experimental website called OUC newsreaders’ website. Finally, we compared the performance of our system with two baseline methods which are traditional Euclidean distance K-nearest neighbour and Naive Bayesian techniques. The result shows a significant improvement of our method over the baseline methods studied.     

Neural networks for GEFCom2017 probabilistic load forecasting

This report describes the forecasting model which was developed by team “4C” for the global energy forecasting competition 2017 (GEFCom2017), with some modifications added afterwards to improve its accuracy. The model is based on neural networks. Temperature scenarios obtained from historical data are used as inputs to the neural networks in order to create load scenarios, and these load scenarios are then transformed into quantiles. By using a feature selection approach that is based on a stepwise regression technique, a neural network based model is developed for each zone. Furthermore, a dynamic choice of the temperature scenarios is suggested. The feature selection and dynamic choice of the temperature scenarios can improve the quantile scores considerably, resulting in very accurate forecasts among the top teams.     

Performance assessment and optimisation of a large information system by combined customer relationship management and resilience engineering: a mathematical programming approach

ISs and ITs play a critical role in large complex gas corporations. Many factors such as human, organisational and environmental factors affect IS in an organisation. Therefore, investigating ISs success is considered to be a complex problem. Also, because of the competitive business environment and the high amount of information flow in organisations, new issues like resilient ISs and successful customer relationship management (CRM) have emerged. A resilient IS will provide sustainable delivery of information to internal and external customers. This paper presents an integrated approach to enhance and optimise the performance of each component of a large IS based on CRM and resilience engineering (RE) in a gas company. The enhancement of the performance can help ISs to perform business tasks efficiently. The data are collected from standard questionnaires. It is then analysed by data envelopment analysis by selecting the optimal mathematical programming approach. The selected model is validated and verified by principle component analysis method. Finally, CRM and RE factors are identified as influential factors through sensitivity analysis for this particular case study. To the best of our knowledge, this is the first study for performance assessment and optimisation of large IS by combined RE and CRM.     

灰色情景法在智能交通技术发展规划中的应用

文章以智能交通的技术发展规划为研究主题,选择ATIS为研究对象,运用情景分析,结合改良灰色统计法来选择该系统需优先发展的相关技术。通过构建情境、分析样本资料,从多个备选方案中确定了需优先发展的技术,其研究成果可用于国内在智能交通技术规划的参考。     

Data features1

This article attempts to provide a formal framework for a data based inference which explicitly and consistently recognizes the approximate nature of probability models. It is based on the idea that a stochastic model is adequate if samples generated under the model are very much like the sample actually obtained. The formalization is based on the concept of data feature. Examples are given of applying the ideas to different areas of statistics including location-scale models, densities, non-parametric regression, interlaboratory test, auto-regressive processes and the analysis of variance. The four cornerstones of the approach are direct comparison, approximation, weak topologies and parsimony. The approach is contrasted to that of much of conventional statistics many of whose concepts are pathologically discontinuous with respect to the topology of data analysis and common sense.     

基于嵌入式平台的高速CCSDS帧处理设备设计

针对空间链路数据传输速率的提高,设计了一种基于嵌入式PowerPC和FPGA平台的高速、可配置的实时帧处理系统。主要描述了系统的硬件电路,采用了PowerPC作为主处理器进行控制与数据处理;采用FPGA作为外部输入输出接口,并进行CCSDS帧的解析处理。FPGA映射到PowerPC中EMIF一段地址空间,通过总线进行数据交互。FPGA设计采用VHDL语言描述,主要对帧同步机制、RS译码进行设计。PowerPC设计采用Vxworks系统,满足实时性要求。实践证明系统的高速性和高可靠性。     

Exploring necessary conditions in HRM research: Fundamental issues and methodological implications

Although the notion that HRM activities are necessary conditions for achieving certain outcomes is widespread in HRM research, most empirical studies follow an additive sufficiency logic. That is, they analyse whether HRM activities increase an outcome, assuming that they can compensate for one another. However, this does not correspond to the idea of necessity where single HRM activities are required for an outcome to occur and cannot be compensated when absent. We discuss the differences between sufficiency and necessity logics in terms of theory, methodology, and practical relevance. Also, we suggest using a new approach and analysis technique in HRM research: necessary condition analysis. We illustrate the application of the method by analysing data on the relationship between ability‐, motivation‐, and opportunity‐enhancing HRM practices (i.e., high‐performance work practices) and employee performance. This illustrative example highlights that necessary conditions require particular theoretical attention and suitable empirical methods. The paper concludes by showing the value of studying necessary conditions, because such analyses allow straightforward recommendations to be given of high relevance for HRM practice, which differ from and add to recommendations based on additive sufficiency logic and analyses.     

Robust multivariate analysis for mixed-type data: Novel algorithm and its practical application in socio-economic research

We propose a novel method and algorithm for the analysis and clustering of mixed-type data using a hierarchical approach based on Forward Search. In our procedure, the identification of groups is based on the identification of similar trajectories and then linked to very intuitive two-dimensional maps. The proposed algorithm can use different measures for the calculation of distance in the case of mixed-type data, such as Gower’s metric and Related metric scaling. A key feature of our algorithm is its ability to discard redundant information from a given set of variables. The practical usefulness of the algorithm is illustrated through two applications of high relevance for empirical economic research. The first one focuses on comparing different indicators of environmental policy stringency in different countries. The second one applies our procedure to identify clusters of countries based on information regarding their institutional characteristics.     

MLP ensembles improve long term prediction accuracy over single networks

This work describes an award winning approach for solving the NN3 Forecasting Competition problem, focusing on the sound experimental validation of its main innovative feature. The NN3 forecasting task consisted of predicting 18 future values of 111 short monthly time series. The main feature of the approach was the use of the median for combining the forecasts of an ensemble of 15 MLPs to predict each time series. Experimental comparison to a single MLP shows that the ensemble increases the performance accuracy for multiple-step ahead forecasting. This system performed well on the withheld data, having finished as the second best solution of the competition with an SMAPE of 16.17%.     

收益还原法在物业维修经济分析中的应用探讨

物业维修是人们越来越关注的领域，其中的经济性分析一直没有一种较好的方法来解决。收益还原法在物业估价中的应用已比较成熟，其原理在物业维修经济分析中有较好的适用性，且基于其原理的物业维修经济分析法也有较好的可操作性。     

A modified GMDH approach for social science research: exploring patterns of relationships in the data

Group Method of Data Handling (GMDH) is a way with which a system of models self-organize themselves by forming higher-order polynomials and selecting the ones with best power of prediction by certain criterion. This method is helpful when we explore patterns of relationships in the data under investigation. In this paper the author presents a modified version of the GMDH algorithm emphasizing the parsimony of models and the behavior of individual parameter estimates as well as of the whole model, and utilizing the consistency and accuracy of bootstrap estimates. This approach is suitable for most research social scientists conduct. An example, the 1907 Romanian Peasant Rebellion, is used to illustrate how to employ the GMDH algorithm when the research topic has been theory-laden. The findings show that GMDH is an appropriate method that social scientists can utilize in their pursuit of a model that is most parsimonious and theoretically meaningful at the same time. Possible extensions of the modified approach, which in its present form works on linear regression type of models, to logit and probit models are also considered.     

跨国公司货运代理商选择模型实证分析

货运代理商的选择合适与否直接影响企业物流业务的顺利完成。针对某跨国公司国际物流影响因素多的特点,从实用角度出发,充分发挥企业物流专家的经验,提出了货运代理商的加权评分法,并且设计了完整的评价指标体系,引入AHP法确定各指标的权重。实例分析结果表示,该方法比直接评分法更能体现企业选择货运代理商的差异性要求。     

Forecasting exchange rates using feedforward and recurrent neural networks

In this paper we investigate the out-of-sample forecasting ability of feedforward and recurrent neural networks based on empirical foreign exchange rate data. A two-step procedure is proposed to construct suitable networks, in which networks are selected based on the predictive stochastic complexity (PSC) criterion, and the selected networks are estimated using both recursive Newton algorithms and the method of nonlinear least squares. Our results show that PSC is a sensible criterion for selecting networks and for certain exchange rate series, some selected network models have significant market timing ability and/or significantly lower out-of-sample mean squared prediction error relative to the random walk model.     

A policy-based rationalization of collective rules: Dimensionality,specialized houses,and decentralized authority

We offer a policy-basis for interpreting, justifying, and designing (3, 3)-political rules, a large class of collective rules analogous to those governing the selection of papers in peer-reviewed journals, where each referee chooses to accept, reject, or invite a resubmission of a paper, and an editor aggregates his own and referees’ opinions into one of these three recommendations. We prove that any such rule is a weighted multicameral rule: a policy is collectively approved at a given level if and only if it is approved by a minimal number of chambers — the dimension of the rule — where each chamber evaluates a different aspect of the policy using a weighted rule, with each evaluator’s weight or authority possibly varying across chambers depending on his area(s) of expertise. These results imply that a given rule is only suitable for evaluating finite-dimensional policies whose dimension corresponds to that of the rule, and they provide a rationale for using different rules to pass different policies even within the same organization. We further introduce the concept of compatibility with a rule and exploit its topological properties to propose a method to construct integer weights corresponding to evaluators’ possible judgments under a given rule, which are more intuitive and easier to interpret for policymakers. Our findings shed light on multicameralism in political institutions and multi-criteria group decision-making in the firm. We provide applications to peer review politics, rating systems, and real-world organizations.     

Using nonparametric methods in social surveys: an empirical study

The most common form of data for socio-economic studies comes from survey sampling. Often the designs of such surveys are complex and use stratification as a method for selecting sample units. A parametric regression model is widely employed for the analysis of such survey data. However the use of a parametric model to represent the relationship between the variables can be inappropriate. A natural alternative is to adopt a nonparametric approach. In this article we address the problem of estimating the finite population mean under stratified sampling. A new stratified estimator based on nonparametric regression is proposed for stratification with proportional allocation, optimum allocation and post-stratification. We focus on an educational and labor-related context with natural populations to test the proposed nonparametric method. Simulated populations have also been considered to evaluate the practical performance of the proposed method.     

How can manufacturers make decisions on product appearance design? A research on optimal design based on customers’ emotional satisfaction

With changing customer attitudes toward consumption and function homogenization, product appearance designs have an increasing influence on the purchase decision. Customer characteristics and emotional factors play an important role here. This study proposes a novel approach for modelling satisfaction and accomplishing a configuration that overcomes the limitations of conventional methods to precisely predict satisfaction, provide optimal product recommendations, and advise manufacturers on product appearance design. The newly proposed approach considers satisfaction, clusters customers through the Kansei perspective, and constructs a satisfaction model for each cluster. Additionally, the study employs data mining to understand the basic design principles and conflicted combinations that must be followed and avoided, respectively. The bidirectional association rules-constrained genetic algorithm is presented to limit configuration freedom, ensuring that results are in the range of control. Comparing prediction errors and recommended sample votes between the novel and conventional approaches revealed the presented approach’s efficiency and accuracy, thereby providing suggestions for manufacturers to make precise decisions on launching new product appearance designs through predicting customer emotional satisfaction.     

Efficiency analysis of organized industrial zones in Eastern Black Sea Region of Turkey

Organized Industrial Zones (OIZs) are the production areas of goods and services that are established to provide planned industrialization and planned urbanization by structuring the industry in suitable areas, to prevent environmental problems, and to provide efficient use of resources. The aim of this study is to propose a decision-making model based on data envelopment analysis (DEA) to evaluate the relative efficiencies of OIZs located in the Eastern Black Sea Region of Turkey. First, the efficiency score of each alternative is determined by employing a DEA formulation with interval data. Second, a common weight DEA-based formulation is applied in order to obtain common set of weights and provide ranking results with an improved discriminating power in imprecise nature. In this study, the best performing OIZ alternative, which performs in Eastern Black Sea region of Turkey, is identified based on the approach proposed by Salahi et al. [1]. The DEA-based model developed by Salahi et al. [1] is improved by modifying a constraint. In order to demonstrate the robustness of the application, two numerical illustrations are given. The first example compares the results obtained by the formulation addressed in Salahi et al. [1] and the improved model; while second numerical illustration provides a case study conducted in Eastern Black Sea region of Turkey.     

Healthcare information systems: data mining methods in the creation of a clinical recommender system

Recommender systems have been extensively studied to present items, such as movies, music and books that are likely of interest to the user. Researchers have indicated that integrated medical information systems are becoming an essential part of the modern healthcare systems. Such systems have evolved to an integrated enterprise-wide system. In particular, such systems are considered as a type of enterprise information systems or ERP system addressing healthcare industry sector needs. As part of efforts, nursing care plan recommender systems can provide clinical decision support, nursing education, clinical quality control, and serve as a complement to existing practice guidelines. We propose to use correlations among nursing diagnoses, outcomes and interventions to create a recommender system for constructing nursing care plans. In the current study, we used nursing diagnosis data to develop the methodology. Our system utilises a prefix-tree structure common in itemset mining to construct a ranked list of suggested care plan items based on previously-entered items. Unlike common commercial systems, our system makes sequential recommendations based on user interaction, modifying a ranked list of suggested items at each step in care plan construction. We rank items based on traditional association-rule measures such as support and confidence, as well as a novel measure that anticipates which selections might improve the quality of future rankings. Since the multi-step nature of our recommendations presents problems for traditional evaluation measures, we also present a new evaluation method based on average ranking position and use it to test the effectiveness of different recommendation strategies.     

A resilience-based model for performance evaluation of information systems: the case of a gas company

Information systems (IS) are strongly influenced by changes in new technology and should react swiftly in response to external conditions. Resilience engineering is a new method that can enable these systems to absorb changes. In this study, a new framework is presented for performance evaluation of IS that includes DeLone and McLean’s factors of success in addition to resilience. Hence, this study is an attempt to evaluate the impact of resilience on IS by the proposed model in Iranian Gas Engineering and Development Company via the data obtained from questionnaires and Fuzzy Data Envelopment Analysis (FDEA) approach. First, FDEA model with α-cut = 0.05 was identified as the most suitable model to this application by performing all Banker, Charnes and Cooper and Charnes, Cooper and Rhodes models of and FDEA and selecting the appropriate model based on maximum mean efficiency. Then, the factors were ranked based on the results of sensitivity analysis, which showed resilience had a significantly higher impact on the proposed model relative to other factors. The results of this study were then verified by conducting the related ANOVA test. This is the first study that examines the impact of resilience on IS by statistical and mathematical approaches.