期刊:
INTERNATIONAL JOURNAL OF INFORMATION TECHNOLOGY & DECISION MAKING,2017年16(3):851-863 ISSN:0219-6220
通讯作者:
Hu, Yi
作者机构:
[Xiao, Yi] Cent China Normal Univ, Sch Informat Management, Wuhan 430079, Peoples R China.;[Wang, Shouyang] Chinese Acad Sci, Acad Math & Syst Sci, Beijing 100190, Peoples R China.;[Xiao, Ming] Cent China Normal Univ, Network Ctr, Wuhan 430079, Peoples R China.;[Xiao, Jin] Sichuan Univ, Business Sch, Chengdu 610064, Peoples R China.;[Hu, Yi] Univ Chinese Acad Sci, Sch Econ & Management, Beijing 100190, Peoples R China.
通讯机构:
[Hu, Yi] U;Univ Chinese Acad Sci, Sch Econ & Management, Beijing 100190, Peoples R China.
关键词:
Cargo volume analyzing;radial basis function network;discrete wavelet technique;TEI@I methodology
期刊:
Industrial Management & Data Systems,2017年117(10):2364-2380 ISSN:0263-5577
通讯作者:
Zhao, Jing
作者机构:
[Wang, Fei; Li, Yajing; Zhao, Jing] China Univ Geosci, Res Ctr Digital Business Management, Sch Econ Management, Wuhan, Hubei, Peoples R China.;[Chi, Maomao] Cent China Normal Univ, Sch Informat Management, Wuhan, Hubei, Peoples R China.
通讯机构:
[Zhao, Jing] C;China Univ Geosci, Res Ctr Digital Business Management, Sch Econ Management, Wuhan, Hubei, Peoples R China.
期刊:
STATISTICS AND ITS INTERFACE,2017年10(1):119-130 ISSN:1938-7989
通讯作者:
Xiao, Yi
作者机构:
[Xiao, Jin] Sichuan Univ, Sch Business, Chengdu 610064, Peoples R China.;[Xu, Lixiang; Xiao, Jin] Univ Munster, Dept Math & Comp Sci, D-48149 Munster, Germany.;[Hu, Yi] Univ Chinese Acad Sci, Sch Econ & Management, Beijing 100190, Peoples R China.;[Xiao, Yi] Cent China Normal Univ, Sch Informat Management, Wuhan 430079, Peoples R China.;[Xu, Lixiang] Hefei Univ, Dept Mathmat & Phys, Hefei 230601, Peoples R China.
通讯机构:
[Xiao, Yi] C;Cent China Normal Univ, Sch Informat Management, Wuhan 430079, Peoples R China.
关键词:
Analog complexing;Crude oil price forecasting;Genetic algorithm;Hybrid transfer learning model;Transfer learning technique
摘要:
Most of the existing models for oil price forecasting only use the data in the forecasted time series. This study proposes a hybrid transfer learning model (HTLM) for crude oil price forecasting. We first selectively transfer some related time series in the source domain to assist in modeling the target time series by using a transfer learning technique, and then construct the forecasting model using the analog complexing (AC) method. Further, we introduce a genetic algorithm to find the optimal match between two important parameters in HTLM. Finally, we use two main crude oil price time series the West Texas Intermediate (WTI) and the Brent crude oil spot prices for empirical analysis. Our results show the effectiveness and superiority of the proposed model compared with existing models.
摘要:
Real-time stream processing is essential for many real-life stream-based applications. Systems designed to run such applications must be prepared to operate under overloaded conditions. In this paper, the load shedding problem is studied for an important class of real-time data stream monitoring applications. In particular, we adopt the deadline model, instead of the commonly used random dropping policy, to capture the QoS requirements of such applications. Based on this model, we propose a Safe lOad Shedding Approach (SOSA) that aims to reduce the workload imposed on the system while at the same time preserve system timing constraints by exploiting data stream semantics. SOSA categorizes stream processing into two different modes and allows one to place provably lighter loads on streams that operate in one particular mode. To demonstrate the usefulness of SOSA, we introduce a concrete scheduling algorithm called SOSA-DBP by combining SOSA with DBP, a well-known scheduling algorithm. Probabilistic analysis and experimental results show that SOSA-DBP has significant performance gain over DBP.
摘要:
Nonnegative matrix factorization (NMF) has received considerable attention due to its interpretation of observed samples as combinations of different components, and has been successfully used as a clustering method. As an extension of NMF, Symmetric NMF (SNMF) inherits the advantages of NMF. Unlike NMF, however, SNMF takes a nonnegative similarity matrix as an input, and two lower rank nonnegative matrices (H, H-T) are computed as an output to approximate the original similarity matrix. Laplacian regularization has improved the clustering performance of NMF and SNMF. However, Laplacian regularization (LR), as a classic manifold regularization method, suffers some problems because of its weak extrapolating ability. In this paper, we propose a novel variant of SNMF, called Hessian regularization based symmetric nonnegative matrix factorization (HSNMF), for this purpose. In contrast to Laplacian regularization, Hessian regularization fits the data perfectly and extrapolates nicely to unseen data. We conduct extensive experiments on several datasets including text data, gene expression data and HMP (Human Microbiome Project) data. The results show that the proposed method outperforms other methods, which suggests the potential application of HSNMF in biological data clustering. (C) 2016 Published by Elsevier Inc.
期刊:
INTERNATIONAL JOURNAL OF DATA MINING AND BIOINFORMATICS,2016年15(2):125-144 ISSN:1748-5673
通讯作者:
Guo, Xiyue
作者机构:
[Guo, Xiyue] Cent China Normal Univ, Natl Engn Res Ctr E Learning, Wuhan, Peoples R China.;[Guo, Xiyue] Xingyi Normal Univ Nationalities, Sch Informat Technol, Xingyi, Peoples R China.;[He, Tingting] Cent China Normal Univ, Sch Comp, Nat Language Proc Lab, Wuhan, Peoples R China.;[Xing, Ying] Zhongyuan Univ Technol, Software Coll, Zhengzhou, Peoples R China.
通讯机构:
[Guo, Xiyue] C;[Guo, Xiyue] X;Cent China Normal Univ, Natl Engn Res Ctr E Learning, Wuhan, Peoples R China.;Xingyi Normal Univ Nationalities, Sch Informat Technol, Xingyi, Peoples R China.
关键词:
PPI extraction;weakly supervised;word dictionary construction;rule learning
摘要:
Each method, machine learning-based and rule-based, for extracting PPI (Protein-Protein Interactions) from biomedical literatures has advantages and disadvantages. In order to utilise the superiorities of these methods reasonably, this paper designs a new structure for the relational word dictionary, uses weakly supervised method to find dictionary items and fill them into the PPI relational word dictionary, and presents a method to learn PPI relational rules automatically based on slot-filling principle. Moreover, this method takes the PPI relation instances without apparent relational words into consideration aiming to improve the final performance. We conduct the experiments with five authoritative biomedical PPI corpuses, and discover some distribution features about PPI relational words. Finally, we also compare our method with several recent research achievements, and the results show that the performance of our method is better than the average level among these methods.
作者机构:
[Li, Yanhui; Lei, Chao; Liu, Bailing] Cent China Normal Univ, Sch Informat Management, Wuhan 430079, Peoples R China.;[Zeng, Bing] S China Univ Technol, Sch Software Engn, Guangzhou 510006, Guangdong, Peoples R China.
通讯机构:
[Zeng, Bing] S;S China Univ Technol, Sch Software Engn, Guangzhou 510006, Guangdong, Peoples R China.
关键词:
automated trust negotiation;mobile commerce;negotiation strategy;Petri net
摘要:
Automated trust negotiation (ATN) offers an attractive means for trust establishments, which establishes mutual trust among strangers wishing to share resources or conduct business, but it comes at the cost of non-trivial computation and communication overheads. The deployment of ATN strategies on a resource-constrained mobile device may lead to user-obstructive latency for operations. In this paper, we propose a trust negotiation strategy called trust target Petri nets negotiation strategy (TPNNS). It highly reduces the negotiation latency in the mobile device compared with other negotiation strategies, since it considers all the alternative responses at each step and chooses the best one. TPNNS supports cycle avoidance and employs skipped TPN which is a new approach presented in this paper. What is more, it is complete and ensures no irrelevant credentials are disclosed during the trust negotiation.
关键词:
ultrafine particles;meteorological parameters;trace gases;industrial city;central China
摘要:
Ultrafine particles with a diameter below 1 μm are strongly linked to traffic and industrial emissions, causing a growing global health concern. In order to reveal the characteristics of ultrafine particles in central China, which makes up the sparse research in industrial cities of a developing country, particle number concentrations (PNC) together with meteorological parameters and concentrations of trace gases were measured over one year inWuhan. The number concentration of ultrafine particles peaked in winter and was the lowest in summer across the entire size range monitored. Further, particles with a diameter smaller than 30 nm increased dramatically in concentration with decreasing diameter. The monthly averaged number concentrations of particles discriminated in three size ranges formed a near- inverse parabolic distribution peaking in January. This trend is supported by a negative correlation between PNC and precipitation, temperature, and mixing layer height, which emphasizes the effect of these meteorological parameters on scouring, convection, and diffusion of particles. However, since wind not only disperses particulate matter but also brings in exogenous particles, wind speed plays an equivocal role in particle number concentrations. The diurnal analysis indicates that hourly measurements of trace gases concentrations could be used as a proxy for dense industrial activities and to reveal some complex chemical reactions. The results of this study offer reasonable estimations of particle impacts and provide references for policymaking of emission control in the industrial cities of developing countries.
摘要:
This paper presents a closed-loop location-inventory-routing problem model considering both quality defect returns and nondefect returns in e-commerce supply chain system. The objective is to minimize the total cost produced in both forward and reverse logistics networks. We propose a combined optimization algorithm named hybrid ant colony optimization algorithm (HACO) to address this model that is an NP-hard problem. Our experimental results show that the proposed HACO is considerably efficient and effective in solving this model.
作者:
Zhang, Jin*;Zhai, Shanshan*;Liu, Hongxia*;Stevenson, Jennifer Ann*
期刊:
JOURNAL OF THE ASSOCIATION FOR INFORMATION SCIENCE AND TECHNOLOGY,2016年67(5):1068-1088 ISSN:2330-1635
通讯作者:
Zhang, Jin;Stevenson, Jennifer Ann;Zhai, Shanshan;Liu, Hongxia
作者机构:
[Zhang, Jin; Zhang, J; Stevenson, Jennifer Ann] Univ Wisconsin, Sch Informat Studies, 2025 E Newport Ave, Milwaukee, WI 53211 USA.;[Zhai, Shanshan] Cent China Normal Univ, Sch Informat Management, 152 Luoyu St, Wuhan 430079, Hu Bei Province, Peoples R China.;[Liu, Hongxia] Ren Min Univ China, Sch Informat Resource Management, 59 Zhongguancun St, Beijing 100872, Peoples R China.
通讯机构:
[Zhang, J; Stevenson, JA] U;[Zhai, Shanshan] C;[Liu, Hongxia] R;Univ Wisconsin, Sch Informat Studies, 2025 E Newport Ave, Milwaukee, WI 53211 USA.;Cent China Normal Univ, Sch Informat Management, 152 Luoyu St, Wuhan 430079, Hu Bei Province, Peoples R China.
关键词:
navigation (information retrieval);portals
摘要:
We investigated a topic-based navigation guidance system in the World Health Organization portal, compared the link connection network and the semantic connection network derived from the guidance system, analyzed the characteristics of the 2 networks from the perspective of the node centrality (in_closeness, out_closeness, betweenness, in_degree, and out_degree), and provided the suggestions to optimize and enhance the topic-based navigation guidance system. A mixed research method that combines the social network analysis method, clustering analysis method, and inferential analysis methods was used. The clustering analysis results of the link connection network were quite different from those of the semantic connection network. There were significant differences between the link connection network and the semantic network in terms of density and centrality. Inferential analysis results show that there were no strong correlations between the centrality of a node and its topic information characteristics. Suggestions for enhancing the navigation guidance system are discussed in detail. Future research directions, such as application of the same research method presented in this study to other similar public health portals, are also included.
作者:
Xiao, Yi;Liu, Yang;Liu, John J.;Xiao, Jin;Hu, Yi*
期刊:
Transportmetrica A: Transport Science,2016年12(1):65-79 ISSN:2324-9935
通讯作者:
Hu, Yi
作者机构:
[Liu, Yang; Xiao, Yi] Cent China Normal Univ, Sch Informat Management, Wuhan 430079, Peoples R China.;[Liu, John J.] City Univ Hong Kong, Ctr Transport Trade & Financial Studies, Hong Kong, Hong Kong, Peoples R China.;[Xiao, Jin] Sichuan Univ, Sch Business, Chengdu 610064, Peoples R China.;[Hu, Yi] Univ Chinese Acad Sci, Sch Econ & Management, Wuhan 100190, Peoples R China.
通讯机构:
[Hu, Yi] U;Univ Chinese Acad Sci, Sch Econ & Management, Wuhan 100190, Peoples R China.
关键词:
passenger flow management;oscillations extracting;operation of airports;forecasting and decision-making
摘要:
Managing air passenger traffic flows is important in investing and operation of airports. However, it is extremely difficult for traditional methods to analyse passenger traffic in both the short and medium terms because of the oscillation and irregularity inherent in air passenger traffic flows dynamics. In this study, we design a hybrid oscillations analysis approach. The proposed method decomposes time series into different scales, making it useful in revealing structural breaks and volatility clusters, and identifying dynamic properties of a process at specific timescales. A case study of Hong Kong airport demonstrates and validates the feasibility of applying the proposed models. Empirical results have confirmed that the proposed model is superior to other competing models and can provide high flexibility in decision-making.
期刊:
European Journal of Operational Research,2016年250(2):521-530 ISSN:0377-2217
通讯作者:
Dong, Qingxing
作者机构:
[Dong, Qingxing] Cent China Normal Univ, Sch Informat Management, Wuhan 430079, Peoples R China.;[Cooper, Orrin] Univ Memphis, Fogelman Coll Business & Econ, Memphis, TN 38152 USA.
通讯机构:
[Dong, Qingxing] C;Cent China Normal Univ, Sch Informat Management, Wuhan 430079, Peoples R China.
关键词:
Consensus reaching;Group decision making;The Analytic Hierarchy Process (AHP);Weight determination
摘要:
Consensus reaching models are widely applied in group decision making problems to improve the group's consensus level before making a common decision. Within the context of the group Analytic Hierarchy Process (AHP), a novel consensus reaching model in a dynamic decision environment is proposed. A Markov chain method can be used to determine the decision makers' weights of importance for the aggregation process with respect to the group members' opinion transition probabilities. The proposed group consensus reaching model facilitates a peer to peer opinion exchange process which relieves the group of the need for a moderator by using an automatic feedback mechanism. Moreover, as the elements in the group decision framework change in a dynamic decision making problem, this model provides feedback suggestions that adaptively adjust for each of the decision makers depending on his credibility in each round. The full process of the dynamic adaptive consensus reaching model is presented and its properties are discussed. Finally, a numerical example is given to demonstrate the effectiveness of our model.
摘要:
In the current business environment, both managers and researchers have realized that assessing and managing risk in a supply chain operation is crucial to business success. Furthermore, the traditional assessment methodologies are unable to deal with intangible criteria which are crucial factor in the analysis. Thus, we develop an orders-of-magnitude AHP (OM-AHP) based ex-ante supply chain risk assessment model, to enable the comparison of the tangible and intangible elements that influence supply chain risks. In the application of OM-AHP method to risk assessment it also became apparent a formal guiding structure of how to pivot using OM-AHP did not exist. A formal method is proposed that can significantly reduce the number of needed comparisons and improve the consistency with pairwise comparisons matrices under any AHP decision. The process of the proposed supply chain risk assessment framework consists of three phases: risk identification, risk assessment, and risk ranking and analysis. An illustrative example is provided to demonstrate the efficacy of the proposed risk assessment framework. The results are organized in a 2-way risk matrix based on their probability and consequence severity and tested for robustness via sensitivity analysis.(C) 2016 Elsevier B.V. All rights reserved.
作者机构:
[Gan, Chunmei] Sun Yat Sen Univ, Sch Informat Management, Guangzhou 510006, Guangdong, Peoples R China.;[Wang, Weijun] Cent China Normal Univ, Sch Informat Management, Wuhan 430079, Peoples R China.;[Gan, Chunmei] Sun Yat Sen Univ, Sch Informat Management, 132 Waihuandong Rd, Guangzhou 510006, Guangdong, Peoples R China.
通讯机构:
[Gan, Chunmei] S;Sun Yat Sen Univ, Sch Informat Management, 132 Waihuandong Rd, Guangzhou 510006, Guangdong, Peoples R China.
关键词:
Social media;Co-word analysis;Bibliometric analysis;Research status;Research trend;China
摘要:
This study aims to map the intellectual structure of social media research in China from 2006 to 2013. Bibliometric and co-word analysis were employed to reveal the characteristics and status on social media research in China. Data was collected from China Academic Journals Full-text Database during the period of 2006---2013. In bibliometric analysis, descriptors of years, themes, subjects, institutions and authors were applied to obtain the research characteristics of social media. In co-word analysis, hierarchical cluster analysis, strategic diagram and social network analysis were adopted. Main results show that, a total of 3178 CSSCI papers on social media have risen yearly and exponentially. The most and distinctive themes were microblog, blog, virtual community and social networking site. The most common subject was News and media, followed by Library, information and digital library, Computer software and application. Wuhan University, Renmin University of China and Nanjing University ranked the top three on the most publications. And the distribution of number of authors with different publications obeys power-law distribution. Moreover, the number of keyword frequency obeys power-law distribution. The core keywords include social media, traditional media, Internet, dissemination and user. There are ten research directions on social media in China, some of which are highly correlated. Generally, the relatively dispersive distribution of research topics suggests the imbalanced development on social media research in China. Some hot topics are well-developed and tend to be mature, a few topics have a great potential for further development, and many other topics are marginal and immature. This study aims to map the intellectual structure of social media research in China from 2006 to 2013. Bibliometric and co-word analysis were employed to reveal the characteristics and status on social media research in China. Data was collected from China Academic Journals Full-text Database during the period of 2006---2013. In bibliometric analysis, descriptors of years, themes, subjects, institutions and authors were applied to obtain the research characteristics of social media. In co-word analysis, hierarchical cluster analysis, strategic diagram and social network analysis were adopted. Main results show that, a total of 3178 CSSCI papers on social media have risen yearly and exponentially. The most and distinctive themes were microblog, blog, virtual community and social networking site. The most common subject was News and media, followed by Library, information and digital library, Computer software and application. Wuhan University, Renmin University of China and Nanjing University ranked the top three on the most publications. And the distribution of number of authors with different publications obeys power-law distribution. Moreover, the number of keyword frequency obeys power-law distribution. The core keywords include social media, traditional media, Internet, dissemination and user. There are ten research directions on social media in China, some of which are highly correlated. Generally, the relatively dispersive distribution of research topics suggests the imbalanced development on social media research in China. Some hot topics are well-developed and tend to be mature, a few topics have a great potential for further development, and many other topics are marginal and immature.
摘要:
Customer churn prediction is one of the key steps to maximize the value of customers for an enterprise. It is difficult to get satisfactory prediction effect by traditional models constructed on the assumption that the training and test data are subject to the same distribution, because the customers usually come from different districts and may be subject to different distributions in reality. This study proposes a feature-selection-based dynamic transfer ensemble (FSDTE) model that aims to introduce transfer learning theory for utilizing the customer data in both the target and related source domains. The model mainly conducts a two-layer feature selection. In the first layer, an initial feature subset is selected by GMDH-type neural network only in the target domain. In the second layer, several appropriate patterns from the source domain to target training set are selected, and some features with higher mutual information between them and the class variable are combined with the initial subset to construct a new feature subset. The selection in the second layer is repeated several times to generate a series of new feature subsets, and then, we train a base classifier in each one. Finally, a best base classifier is selected dynamically for each test pattern. The experimental results in two customer churn prediction datasets show that FSDTE can achieve better performance compared with the traditional churn prediction strategies, as well as three existing transfer learning strategies.