The feature selection in analogy-based software effort estimation （ASEE） is formulized as a multi-objective optimization problem. One objective is designed to maximize the effort estimation accuracy and the other ob...The feature selection in analogy-based software effort estimation （ASEE） is formulized as a multi-objective optimization problem. One objective is designed to maximize the effort estimation accuracy and the other objective is designed to minimize the number of selected features. Based on these two potential conflict objectives, a novel wrapper- based feature selection method, multi-objective feature selection for analogy-based software effort estimation （MASE）, is proposed. In the empirical studies, 77 projects in Desharnais and 62 projects in Maxwell from the real world are selected as the evaluation objects and the proposed method MASE is compared with some baseline methods. Final results show that the proposed method can achieve better performance by selecting fewer features when considering MMRE （mean magnitude of relative error）, MdMRE （median magnitude of relative error）, PRED （ 0. 25 ）, and SA （ standardized accuracy） performance metrics.展开更多
Deep convolutional neural networks(CNNs)have demonstrated remarkable performance in video super-resolution(VSR).However,the ability of most existing methods to recover fine details in complex scenes is often hindered ...Deep convolutional neural networks(CNNs)have demonstrated remarkable performance in video super-resolution(VSR).However,the ability of most existing methods to recover fine details in complex scenes is often hindered by the loss of shallow texture information during feature extraction.To address this limitation,we propose a 3D Convolutional Enhanced Residual Video Super-Resolution Network(3D-ERVSNet).This network employs a forward and backward bidirectional propagation module(FBBPM)that aligns features across frames using explicit optical flow through lightweight SPyNet.By incorporating an enhanced residual structure(ERS)with skip connections,shallow and deep features are effectively integrated,enhancing texture restoration capabilities.Furthermore,3D convolution module(3DCM)is applied after the backward propagation module to implicitly capture spatio-temporal dependencies.The architecture synergizes these components where FBBPM extracts aligned features,ERS fuses hierarchical representations,and 3DCM refines temporal coherence.Finally,a deep feature aggregation module(DFAM)fuses the processed features,and a pixel-upsampling module(PUM)reconstructs the high-resolution(HR)video frames.Comprehensive evaluations on REDS,Vid4,UDM10,and Vim4 benchmarks demonstrate well performance including 30.95 dB PSNR/0.8822 SSIM on REDS and 32.78 dB/0.8987 on Vim4.3D-ERVSNet achieves significant gains over baselines while maintaining high efficiency with only 6.3M parameters and 77ms/frame runtime(i.e.,20×faster than RBPN).The network’s effectiveness stems from its task-specific asymmetric design that balances explicit alignment and implicit fusion.展开更多
Point of interest(POI)recommendation analyses user preferences through historical check-in data.However,existing POI recommendation methods often overlook the influence of weather information and face the challenge of...Point of interest(POI)recommendation analyses user preferences through historical check-in data.However,existing POI recommendation methods often overlook the influence of weather information and face the challenge of sparse historical data for individual users.To address these issues,this paper proposes a new paradigm,namely temporal-weather-aware transition pattern for POI recommendation(TWTransNet).This paradigm is designed to capture user transition patterns under different times and weather conditions.Additionally,we introduce the construction of a user-POI interaction graph to alleviate the problem of sparse historical data for individual users.Furthermore,when predicting user interests by aggregating graph information,some POIs may not be suitable for visitation under current weather conditions.To account for this,we propose an attention mechanism to filter POI neighbours when aggregating information from the graph,considering the impact of weather and time.Empirical results on two real-world datasets demonstrate the superior performance of our proposed method,showing a substantial improvement of 6.91%-23.31% in terms of prediction accuracy.展开更多
Recurrent neural networks （RNN） have been very successful in handling sequence data. However, understanding RNN and finding the best practices for RNN learning is a difficult task, partly because there are many comp...Recurrent neural networks （RNN） have been very successful in handling sequence data. However, understanding RNN and finding the best practices for RNN learning is a difficult task, partly because there are many competing and complex hidden units, such as the long short-term memory （LSTM） and the gated recurrent unit （GRU）. We propose a gated unit for RNN, named as minimal gated unit （MCU）, since it only contains one gate, which is a minimal design among all gated hidden units. The design of MCU benefits from evaluation results on LSTM and GRU in the literature. Experiments on various sequence data show that MCU has comparable accuracy with GRU, but has a simpler structure, fewer parameters, and faster training. Hence, MGU is suitable in RNN＇s applications. Its simple architecture also means that it is easier to evaluate and tune, and in principle it is easier to study MGU＇s properties theoretically and empirically.展开更多
During the past few decades,mobile wireless communications have experienced four generations of technological revolution,namely from 1 G to 4 G,and the deployment of the latest 5 G networks is expected to take place i...During the past few decades,mobile wireless communications have experienced four generations of technological revolution,namely from 1 G to 4 G,and the deployment of the latest 5 G networks is expected to take place in 2019.One fundamental question is how we can push forward the development of mobile wireless communications while it has become an extremely complex and sophisticated system.We believe that the answer lies in the huge volumes of data produced by the network itself,and machine learning may become a key to exploit such information.In this paper,we elaborate why the conventional model-based paradigm,which has been widely proved useful in pre-5 G networks,can be less efficient or even less practical in the future 5 G and beyond mobile networks.Then,we explain how the data-driven paradigm,using state-of-the-art machine learning techniques,can become a promising solution.At last,we provide a typical use case of the data-driven paradigm,i.e.,proactive load balancing,in which online learning is utilized to adjust cell configurations in advance to avoid burst congestion caused by rapid traffic changes.展开更多
The cloud storage service cannot be completely trusted because of the separation of data management and ownership, leading to the difficulty of data privacy protection. In order to protect the privacy of data on untru...The cloud storage service cannot be completely trusted because of the separation of data management and ownership, leading to the difficulty of data privacy protection. In order to protect the privacy of data on untrusted servers of cloud storage, a novel multi-authority access control scheme without a trustworthy central authority has been proposed based on CP-ABE for cloud storage systems, called non-centered multi-authority proxy re-encryption based on the cipher-text policy attribute-based encryption(NC-MACPABE). NC-MACPABE optimizes the weighted access structure(WAS) allowing different levels of operation on the same file in cloud storage system. The concept of identity dyeing is introduced to improve the users' information privacy further. The re-encryption algorithm is improved in the scheme so that the data owner can revoke user's access right in a more flexible way. The scheme is proved to be secure. And the experimental results also show that removing the central authority can resolve the existing performance bottleneck in the multi-authority architecture with a central authority, which significantly improves user experience when a large number of users apply for accesses to the cloud storage system at the same time.展开更多
The Cloud is increasingly being used to store and process big data for its tenants and classical security mechanisms using encryption are neither sufficiently efficient nor suited to the task of protecting big data in...The Cloud is increasingly being used to store and process big data for its tenants and classical security mechanisms using encryption are neither sufficiently efficient nor suited to the task of protecting big data in the Cloud.In this paper,we present an alternative approach which divides big data into sequenced parts and stores them among multiple Cloud storage service providers.Instead of protecting the big data itself,the proposed scheme protects the mapping of the various data elements to each provider using a trapdoor function.Analysis,comparison and simulation prove that the proposed scheme is efficient and secure for the big data of Cloud tenants.展开更多
In a measurement system, new representation methods are necessary to maintain the uncertainty and to supply more powerful ability for reasoning and transformation between numerical system and symbolic system. A grey m...In a measurement system, new representation methods are necessary to maintain the uncertainty and to supply more powerful ability for reasoning and transformation between numerical system and symbolic system. A grey measurement system is discussed from the point of view of intelligent sensors and incomplete information processing compared with a numerical and symbolized measurement system. The methods of grey representation and information processing are proposed for data collection and reasoning. As a case study, multi-ultrasonic sensor systems are demonstrated to verify the effectiveness of the proposed methods.展开更多
How to effectively reduce the energy consumption of large-scale data centers is a key issue in cloud computing. This paper presents a novel low-power task scheduling algorithm （L3SA） for large-scale cloud data cente...How to effectively reduce the energy consumption of large-scale data centers is a key issue in cloud computing. This paper presents a novel low-power task scheduling algorithm （L3SA） for large-scale cloud data centers. The winner tree is introduced to make the data nodes as the leaf nodes of the tree and the final winner on the purpose of reducing energy consumption is selected. The complexity of large-scale cloud data centers is fully consider, and the task comparson coefficient is defined to make task scheduling strategy more reasonable. Experiments and performance analysis show that the proposed algorithm can effectively improve the node utilization, and reduce the overall power consumption of the cloud data center.展开更多
Achieving good recognition results for License plates is challenging due to multiple adverse factors. For instance, in Malaysia, where private vehicle （e.g., cars） have numbers with dark background, while public veh...Achieving good recognition results for License plates is challenging due to multiple adverse factors. For instance, in Malaysia, where private vehicle （e.g., cars） have numbers with dark background, while public vehicle （taxis/cabs） have numbers with white background. To reduce the complexity of the problem, we propose to classify the above two types of images such that one can choose an appropriate method to achieve better results. Therefore, in this work, we explore the combination of Convolutional Neural Networks （CNN） and Recurrent Neural Networks namely, BLSTM （Bi-Directional Long Short Term Memory）, for recognition. The CNN has been used for feature extraction as it has high discriminative ability, at the same time, BLSTM has the ability to extract context information based on the past information. For classification, we propose Dense Cluster based Voting （DCV）, which separates foreground and background for successful classification of private and public. Experimental results on live data given by MIMOS, which is funded by Malaysian Government and the standard dataset UCSD show that the proposed classification outperforms the existing methods. In addition, the recognition results show that the recognition performance improves significantly after classification compared to before classification.展开更多
Reinforcement learning(RL)algorithm has been introduced for several decades,which becomes a paradigm in sequential decision-making and control.The development of reinforcement learning,especially in recent years,has e...Reinforcement learning(RL)algorithm has been introduced for several decades,which becomes a paradigm in sequential decision-making and control.The development of reinforcement learning,especially in recent years,has enabled this algorithm to be applied in many industry fields,such as robotics,medical intelligence,and games.This paper first introduces the history and background of reinforcement learning,and then illustrates the industrial application and open source platforms.After that,the successful applications from AlphaGo to AlphaZero and future reinforcement learning technique are focused on.Finally,the artificial intelligence for complex interaction(e.g.,stochastic environment,multiple players,selfish behavior,and distributes optimization)is considered and this paper concludes with the highlight and outlook of future general artificial intelligence.展开更多
Self-adaptive software is an efficient way to cope with highly dynamic nature of the environment where it is situated. In this paper, from the perspective of software architecture, we propose a component framework for...Self-adaptive software is an efficient way to cope with highly dynamic nature of the environment where it is situated. In this paper, from the perspective of software architecture, we propose a component framework for supporting the architecture-based design and development of self-adaptive applications. It captures some key elements of the research on software architecture and provides more flexible facilities to decouple interacting components. Based on that, a prototype is implemented to demonstrate its feasibility, and at last a case study is presented to illustrate our framework.展开更多
This paper sums up four security factors after analyzing co-residency threats caused by the special multitenant environment in the cloud.To secure the factors,a multiway dynamic trust chain transfer model was proposed...This paper sums up four security factors after analyzing co-residency threats caused by the special multitenant environment in the cloud.To secure the factors,a multiway dynamic trust chain transfer model was proposed on the basis of a measurement interactive virtual machine and current behavior to protect the integrity of the system.A trust chain construction module is designed in a virtual machine monitor.Through dynamic monitoring,it achieves the purpose of transferring integrity between virtual machine.A cloud system with a trust authentication function is implemented on the basis of the model,and its practicability is shown.展开更多
feature representations from a large amount of data,and use reinforcement learning to learn the best strategy to complete the task.Through the combination of deep learning and reinforcement learning,end-to-end input a...feature representations from a large amount of data,and use reinforcement learning to learn the best strategy to complete the task.Through the combination of deep learning and reinforcement learning,end-to-end input and output can be achieved,and substantial breakthroughs have been made in many planning and decision-making systems with infinite states,such as games,in particular,AlphaGo,robotics,natural language processing,dialogue systems,machine translation,and computer vision.In this paper we have summarized the main techniques of deep reinforcement learning and its applications in image processing.展开更多
Word segmentation is an integral step in many knowledge discovery applications. However, existing word segmentation methods have problems when applying to Chinese judicial documents:(1) existing methods rely on large-...Word segmentation is an integral step in many knowledge discovery applications. However, existing word segmentation methods have problems when applying to Chinese judicial documents:(1) existing methods rely on large-scale labeled data which is typically unavailable in judicial documents, and (2) judicial document has its own language features and writing formats. In this paper, a word segmentation method is proposed for Chinese judicial documents. The proposed method consists of two steps:(1) automatically generating some labeled data as legal dictionaries, and (2) applying a hybrid multilayer neural networks to do word segmentation incorporating legal dictionaries. Experiments are conducted on a dataset of Chinese judicial documents showing that the proposed model can achieve better results than the existing methods.展开更多
Multi-objective genetic algorithm is much suitable for solving multi-objective optimization problems. By use of Genetic algorithm, the optimization of S-boxes is explored in this paper. Results of the experiments show...Multi-objective genetic algorithm is much suitable for solving multi-objective optimization problems. By use of Genetic algorithm, the optimization of S-boxes is explored in this paper. Results of the experiments show that, with heuristic mutation strategy, the algorithm has high searching efficiency and fast convergence speed. Meanwhile, we also have take the avalanche probability of S-boxes into account, besides nonlinearity and difference uniformity. Under this method, an effective genetic algorithm for 6×6 S-boxes is provided and a number of S-boxes with good cryptographic capability can be obtained.展开更多
Web service composition lets developers create applications on top of service-oriented computing and its native description, discovery, and communication capabilities. This paper mainly focuses on the QoS when the con...Web service composition lets developers create applications on top of service-oriented computing and its native description, discovery, and communication capabilities. This paper mainly focuses on the QoS when the concrete composition structure is unknown. A QoS model of service composition is presented based on the fuzzy directed graph theory. According to the model, a recursive algorithm is also described for calculating such kind of QoS. And, the feasibility of this QoS model and the recursive algorithm is verified by a case study. The proposed approach enables customers to get a possible value of the QoS before they achieve the service.展开更多
Starting from late 2019,the new coronavirus disease(COVID-19)has become a global crisis.With the development of online social media,people prefer to express their opinions and discuss the latest news online.We have wi...Starting from late 2019,the new coronavirus disease(COVID-19)has become a global crisis.With the development of online social media,people prefer to express their opinions and discuss the latest news online.We have witnessed the positive influence of online social media,which helped citizens and governments track the development of this pandemic in time.It is necessary to apply artificial intelligence(AI)techniques to online social media and automatically discover and track public opinions posted online.In this paper,we take Sina Weibo,the most widely used online social media in China,for analysis and experiments.We collect multi-modal microblogs about COVID-19 from 2020/1/1 to 2020/3/31 with a web crawler,including texts and images posted by users.In order to effectively discover what is being discussed about COVID-19 without human labeling,we propose a unified multi-modal framework,including an unsupervised short-text topic model to discover and track bursty topics,and a self-supervised model to learn image features so that we can retrieve related images about COVID-19.Experimental results have shown the effectiveness and superiority of the proposed models,and also have shown the considerable application prospects for analyzing and tracking public opinions about COVID-19.展开更多
Kantorovich theorem was extended to variational inequalities by which the convergence of Newton iteration,the existence and uniqueness of the solution of the problem can be tested via computational conditions at the i...Kantorovich theorem was extended to variational inequalities by which the convergence of Newton iteration,the existence and uniqueness of the solution of the problem can be tested via computational conditions at the initial point.展开更多
Activity is now playing a vital role in software processes. To ensure the high-level efficiency of software processes, a key point is to locate those activities that own bigger resource occupation probabilities with r...Activity is now playing a vital role in software processes. To ensure the high-level efficiency of software processes, a key point is to locate those activities that own bigger resource occupation probabilities with respect to average execution time, called delayed activities, and then improve them. To this end, we firstly propose an approach to locating delayed activities in software processes. Furthermore, we present a case study, which exhibits the high-level efficiency of the approach, to concretely illustrate this new solution. Some beneficial analysis and reasonable modification are developed in the end.展开更多
基金The National Natural Science Foundation of China(No.61602267,61202006)the Open Project of State Key Laboratory for Novel Software Technology at Nanjing University(No.KFKT2016B18)
文摘The feature selection in analogy-based software effort estimation （ASEE） is formulized as a multi-objective optimization problem. One objective is designed to maximize the effort estimation accuracy and the other objective is designed to minimize the number of selected features. Based on these two potential conflict objectives, a novel wrapper- based feature selection method, multi-objective feature selection for analogy-based software effort estimation （MASE）, is proposed. In the empirical studies, 77 projects in Desharnais and 62 projects in Maxwell from the real world are selected as the evaluation objects and the proposed method MASE is compared with some baseline methods. Final results show that the proposed method can achieve better performance by selecting fewer features when considering MMRE （mean magnitude of relative error）, MdMRE （median magnitude of relative error）, PRED （ 0. 25 ）, and SA （ standardized accuracy） performance metrics.
基金supported in part by the Basic and Applied Basic Research Foundation of Guangdong Province[2025A1515011566]in part by the State Key Laboratory for Novel Software Technology,Nanjing University[KFKT2024B08]+1 种基金in part by Leading Talents in Gusu Innovation and Entrepreneurship[ZXL2023170]in part by the Basic Research Programs of Taicang 2024,[TC2024JC32].
文摘Deep convolutional neural networks(CNNs)have demonstrated remarkable performance in video super-resolution(VSR).However,the ability of most existing methods to recover fine details in complex scenes is often hindered by the loss of shallow texture information during feature extraction.To address this limitation,we propose a 3D Convolutional Enhanced Residual Video Super-Resolution Network(3D-ERVSNet).This network employs a forward and backward bidirectional propagation module(FBBPM)that aligns features across frames using explicit optical flow through lightweight SPyNet.By incorporating an enhanced residual structure(ERS)with skip connections,shallow and deep features are effectively integrated,enhancing texture restoration capabilities.Furthermore,3D convolution module(3DCM)is applied after the backward propagation module to implicitly capture spatio-temporal dependencies.The architecture synergizes these components where FBBPM extracts aligned features,ERS fuses hierarchical representations,and 3DCM refines temporal coherence.Finally,a deep feature aggregation module(DFAM)fuses the processed features,and a pixel-upsampling module(PUM)reconstructs the high-resolution(HR)video frames.Comprehensive evaluations on REDS,Vid4,UDM10,and Vim4 benchmarks demonstrate well performance including 30.95 dB PSNR/0.8822 SSIM on REDS and 32.78 dB/0.8987 on Vim4.3D-ERVSNet achieves significant gains over baselines while maintaining high efficiency with only 6.3M parameters and 77ms/frame runtime(i.e.,20×faster than RBPN).The network’s effectiveness stems from its task-specific asymmetric design that balances explicit alignment and implicit fusion.
基金supported by Stable Support Project of Shenzhen(20231120161634002)Shenzhen Science and Technology Programme(JCYJ20240813141417023)+5 种基金Natural Science Foundation of Guangdong Province of China(2025A1515010233)Guangdong Provincial Department of Education(2024KTSCX060)Tencent‘Rhinoceros Birds’—Scientific Research Foundation for Young Teachers of Shenzhen University,Open Project of State Key Laboratory for Novel Software Technology of Nanjing University(KFKT2025B22)Hong Kong RGC General Research Fund(No.152211/23E and 15216424/24E)PolyU Internal Fund(No.P0043932,P0048988)NVIDIA AI Technology Centre.
文摘Point of interest(POI)recommendation analyses user preferences through historical check-in data.However,existing POI recommendation methods often overlook the influence of weather information and face the challenge of sparse historical data for individual users.To address these issues,this paper proposes a new paradigm,namely temporal-weather-aware transition pattern for POI recommendation(TWTransNet).This paradigm is designed to capture user transition patterns under different times and weather conditions.Additionally,we introduce the construction of a user-POI interaction graph to alleviate the problem of sparse historical data for individual users.Furthermore,when predicting user interests by aggregating graph information,some POIs may not be suitable for visitation under current weather conditions.To account for this,we propose an attention mechanism to filter POI neighbours when aggregating information from the graph,considering the impact of weather and time.Empirical results on two real-world datasets demonstrate the superior performance of our proposed method,showing a substantial improvement of 6.91%-23.31% in terms of prediction accuracy.
基金supported by National Natural Science Foundation of China(Nos.61422203 and 61333014)National Key Basic Research Program of China(No.2014CB340501)
文摘Recurrent neural networks （RNN） have been very successful in handling sequence data. However, understanding RNN and finding the best practices for RNN learning is a difficult task, partly because there are many competing and complex hidden units, such as the long short-term memory （LSTM） and the gated recurrent unit （GRU）. We propose a gated unit for RNN, named as minimal gated unit （MCU）, since it only contains one gate, which is a minimal design among all gated hidden units. The design of MCU benefits from evaluation results on LSTM and GRU in the literature. Experiments on various sequence data show that MCU has comparable accuracy with GRU, but has a simpler structure, fewer parameters, and faster training. Hence, MGU is suitable in RNN＇s applications. Its simple architecture also means that it is easier to evaluate and tune, and in principle it is easier to study MGU＇s properties theoretically and empirically.
基金partially supported by the National Natural Science Foundation of China(61751306,61801208,61671233)the Jiangsu Science Foundation(BK20170650)+2 种基金the Postdoctoral Science Foundation of China(BX201700118,2017M621712)the Jiangsu Postdoctoral Science Foundation(1701118B)the Fundamental Research Funds for the Central Universities(021014380094)
文摘During the past few decades,mobile wireless communications have experienced four generations of technological revolution,namely from 1 G to 4 G,and the deployment of the latest 5 G networks is expected to take place in 2019.One fundamental question is how we can push forward the development of mobile wireless communications while it has become an extremely complex and sophisticated system.We believe that the answer lies in the huge volumes of data produced by the network itself,and machine learning may become a key to exploit such information.In this paper,we elaborate why the conventional model-based paradigm,which has been widely proved useful in pre-5 G networks,can be less efficient or even less practical in the future 5 G and beyond mobile networks.Then,we explain how the data-driven paradigm,using state-of-the-art machine learning techniques,can become a promising solution.At last,we provide a typical use case of the data-driven paradigm,i.e.,proactive load balancing,in which online learning is utilized to adjust cell configurations in advance to avoid burst congestion caused by rapid traffic changes.
基金Projects(61472192,61202004)supported by the National Natural Science Foundation of ChinaProject(14KJB520014)supported by the Natural Science Fund of Higher Education of Jiangsu Province,China
文摘The cloud storage service cannot be completely trusted because of the separation of data management and ownership, leading to the difficulty of data privacy protection. In order to protect the privacy of data on untrusted servers of cloud storage, a novel multi-authority access control scheme without a trustworthy central authority has been proposed based on CP-ABE for cloud storage systems, called non-centered multi-authority proxy re-encryption based on the cipher-text policy attribute-based encryption(NC-MACPABE). NC-MACPABE optimizes the weighted access structure(WAS) allowing different levels of operation on the same file in cloud storage system. The concept of identity dyeing is introduced to improve the users' information privacy further. The re-encryption algorithm is improved in the scheme so that the data owner can revoke user's access right in a more flexible way. The scheme is proved to be secure. And the experimental results also show that removing the central authority can resolve the existing performance bottleneck in the multi-authority architecture with a central authority, which significantly improves user experience when a large number of users apply for accesses to the cloud storage system at the same time.
基金supported in part by the National Nature Science Foundation of China under Grant No.61402413 and 61340058 the "Six Kinds Peak Talents Plan" project of Jiangsu Province under Grant No.ll-JY-009+2 种基金the Nature Science Foundation of Zhejiang Province under Grant No.LY14F020019, Z14F020006 and Y1101183the China Postdoctoral Science Foundation funded project under Grant No.2012M511732Jiangsu Province Postdoctoral Science Foundation funded project Grant No.1102014C
文摘The Cloud is increasingly being used to store and process big data for its tenants and classical security mechanisms using encryption are neither sufficiently efficient nor suited to the task of protecting big data in the Cloud.In this paper,we present an alternative approach which divides big data into sequenced parts and stores them among multiple Cloud storage service providers.Instead of protecting the big data itself,the proposed scheme protects the mapping of the various data elements to each provider using a trapdoor function.Analysis,comparison and simulation prove that the proposed scheme is efficient and secure for the big data of Cloud tenants.
基金the National Natural Science Foundation of China (6070308360575033).
文摘In a measurement system, new representation methods are necessary to maintain the uncertainty and to supply more powerful ability for reasoning and transformation between numerical system and symbolic system. A grey measurement system is discussed from the point of view of intelligent sensors and incomplete information processing compared with a numerical and symbolized measurement system. The methods of grey representation and information processing are proposed for data collection and reasoning. As a case study, multi-ultrasonic sensor systems are demonstrated to verify the effectiveness of the proposed methods.
基金supported by the National Natural Science Foundation of China(6120200461272084)+9 种基金the National Key Basic Research Program of China(973 Program)(2011CB302903)the Specialized Research Fund for the Doctoral Program of Higher Education(2009322312000120113223110003)the China Postdoctoral Science Foundation Funded Project(2011M5000952012T50514)the Natural Science Foundation of Jiangsu Province(BK2011754BK2009426)the Jiangsu Postdoctoral Science Foundation Funded Project(1102103C)the Natural Science Fund of Higher Education of Jiangsu Province(12KJB520007)the Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions(yx002001)
文摘How to effectively reduce the energy consumption of large-scale data centers is a key issue in cloud computing. This paper presents a novel low-power task scheduling algorithm （L3SA） for large-scale cloud data centers. The winner tree is introduced to make the data nodes as the leaf nodes of the tree and the final winner on the purpose of reducing energy consumption is selected. The complexity of large-scale cloud data centers is fully consider, and the task comparson coefficient is defined to make task scheduling strategy more reasonable. Experiments and performance analysis show that the proposed algorithm can effectively improve the node utilization, and reduce the overall power consumption of the cloud data center.
基金This research work was supported by the Faculty of Computer Science and Information Technology, the University of Malaya under a special allocation of Post Graduate Funding for the RP036B-15AET project. The work described in this paper was supported by the Natural Science Foundation of China under grant no. 61672273, and the Science Foundation for Distinguished Young Scholars of Jiangsu under grant no. BK20160021.
文摘Achieving good recognition results for License plates is challenging due to multiple adverse factors. For instance, in Malaysia, where private vehicle （e.g., cars） have numbers with dark background, while public vehicle （taxis/cabs） have numbers with white background. To reduce the complexity of the problem, we propose to classify the above two types of images such that one can choose an appropriate method to achieve better results. Therefore, in this work, we explore the combination of Convolutional Neural Networks （CNN） and Recurrent Neural Networks namely, BLSTM （Bi-Directional Long Short Term Memory）, for recognition. The CNN has been used for feature extraction as it has high discriminative ability, at the same time, BLSTM has the ability to extract context information based on the past information. For classification, we propose Dense Cluster based Voting （DCV）, which separates foreground and background for successful classification of private and public. Experimental results on live data given by MIMOS, which is funded by Malaysian Government and the standard dataset UCSD show that the proposed classification outperforms the existing methods. In addition, the recognition results show that the recognition performance improves significantly after classification compared to before classification.
文摘Reinforcement learning(RL)algorithm has been introduced for several decades,which becomes a paradigm in sequential decision-making and control.The development of reinforcement learning,especially in recent years,has enabled this algorithm to be applied in many industry fields,such as robotics,medical intelligence,and games.This paper first introduces the history and background of reinforcement learning,and then illustrates the industrial application and open source platforms.After that,the successful applications from AlphaGo to AlphaZero and future reinforcement learning technique are focused on.Finally,the artificial intelligence for complex interaction(e.g.,stochastic environment,multiple players,selfish behavior,and distributes optimization)is considered and this paper concludes with the highlight and outlook of future general artificial intelligence.
基金Supported bythe National Key Basic Research andDevelopment Programof China (2002CB312002) the National Natu-ral Science Foundation of China (60403014 ,60233010) +1 种基金the NationalHigh-Tech Research and Development Program of China(2005AA113030) the Natural Science Foundation of Jiangsu Prov-ince (BK2006712)
文摘Self-adaptive software is an efficient way to cope with highly dynamic nature of the environment where it is situated. In this paper, from the perspective of software architecture, we propose a component framework for supporting the architecture-based design and development of self-adaptive applications. It captures some key elements of the research on software architecture and provides more flexible facilities to decouple interacting components. Based on that, a prototype is implemented to demonstrate its feasibility, and at last a case study is presented to illustrate our framework.
基金supported by The National Natural Science Foundation for Young Scientists of China under Grant No.61303263the Jiangsu Provincial Research Foundation for Basic Research(Natural Science Foundation)under Grant No.BK20150201+4 种基金the Scientific Research Key Project of Beijing Municipal Commission of Education under Grant No.KZ201210015015Project Supported by the National Natural Science Foundation of China(Grant No.61370140)the Scientific Research Common Program of the Beijing Municipal Commission of Education(Grant No.KMKM201410015006)The National Science Foundation of China under Grant Nos.61232016 and U1405254and the PAPD fund
文摘This paper sums up four security factors after analyzing co-residency threats caused by the special multitenant environment in the cloud.To secure the factors,a multiway dynamic trust chain transfer model was proposed on the basis of a measurement interactive virtual machine and current behavior to protect the integrity of the system.A trust chain construction module is designed in a virtual machine monitor.Through dynamic monitoring,it achieves the purpose of transferring integrity between virtual machine.A cloud system with a trust authentication function is implemented on the basis of the model,and its practicability is shown.
基金This work was supported in part by the Open Research Project of State Key Laboratory of Novel Software Technology under Grant KFKT2018B23the Priority Academic Program Development of Jiangsu Higher Education Institutions,the 2018 Tiancheng Huizhi Innovation Promotion Education and Scientific Research Innovation Fund of the Ministry of Education under Grant 2018A03038 and the Open Project Program of the State Key Lab of CAD&CG(Grant No.A1916),Zhejiang University.
文摘feature representations from a large amount of data,and use reinforcement learning to learn the best strategy to complete the task.Through the combination of deep learning and reinforcement learning,end-to-end input and output can be achieved,and substantial breakthroughs have been made in many planning and decision-making systems with infinite states,such as games,in particular,AlphaGo,robotics,natural language processing,dialogue systems,machine translation,and computer vision.In this paper we have summarized the main techniques of deep reinforcement learning and its applications in image processing.
文摘Word segmentation is an integral step in many knowledge discovery applications. However, existing word segmentation methods have problems when applying to Chinese judicial documents:(1) existing methods rely on large-scale labeled data which is typically unavailable in judicial documents, and (2) judicial document has its own language features and writing formats. In this paper, a word segmentation method is proposed for Chinese judicial documents. The proposed method consists of two steps:(1) automatically generating some labeled data as legal dictionaries, and (2) applying a hybrid multilayer neural networks to do word segmentation incorporating legal dictionaries. Experiments are conducted on a dataset of Chinese judicial documents showing that the proposed model can achieve better results than the existing methods.
基金Supported by the National Natural Science Foundation of China (60473012)
文摘Multi-objective genetic algorithm is much suitable for solving multi-objective optimization problems. By use of Genetic algorithm, the optimization of S-boxes is explored in this paper. Results of the experiments show that, with heuristic mutation strategy, the algorithm has high searching efficiency and fast convergence speed. Meanwhile, we also have take the avalanche probability of S-boxes into account, besides nonlinearity and difference uniformity. Under this method, an effective genetic algorithm for 6×6 S-boxes is provided and a number of S-boxes with good cryptographic capability can be obtained.
基金Supported by the National Natural Science Foundation of China(60303025 ,60673017)the Natural Science Foundation of Jiangsu Prov-ince (BK2007137)the Program for New Century Excellent Talents in University
文摘Web service composition lets developers create applications on top of service-oriented computing and its native description, discovery, and communication capabilities. This paper mainly focuses on the QoS when the concrete composition structure is unknown. A QoS model of service composition is presented based on the fuzzy directed graph theory. According to the model, a recursive algorithm is also described for calculating such kind of QoS. And, the feasibility of this QoS model and the recursive algorithm is verified by a case study. The proposed approach enables customers to get a possible value of the QoS before they achieve the service.
基金This paper is supported by the Fundamental Research Funds for the Central Universities[No.JUSRP12021].
文摘Starting from late 2019,the new coronavirus disease(COVID-19)has become a global crisis.With the development of online social media,people prefer to express their opinions and discuss the latest news online.We have witnessed the positive influence of online social media,which helped citizens and governments track the development of this pandemic in time.It is necessary to apply artificial intelligence(AI)techniques to online social media and automatically discover and track public opinions posted online.In this paper,we take Sina Weibo,the most widely used online social media in China,for analysis and experiments.We collect multi-modal microblogs about COVID-19 from 2020/1/1 to 2020/3/31 with a web crawler,including texts and images posted by users.In order to effectively discover what is being discussed about COVID-19 without human labeling,we propose a unified multi-modal framework,including an unsupervised short-text topic model to discover and track bursty topics,and a self-supervised model to learn image features so that we can retrieve related images about COVID-19.Experimental results have shown the effectiveness and superiority of the proposed models,and also have shown the considerable application prospects for analyzing and tracking public opinions about COVID-19.
文摘Kantorovich theorem was extended to variational inequalities by which the convergence of Newton iteration,the existence and uniqueness of the solution of the problem can be tested via computational conditions at the initial point.
基金supported by National Natural Science Foundation of China(No.61462091)High-tech Industrial Development Program of Yunnan Province(No.1956,in 2012)+2 种基金New Academic Researcher Award for Doctoral Candidates of Yunnan Province of China(No.ynu201414)Natural Science Youth Foundation of Yunnan Province of China(No.2014FD006)the Postgraduates Science Foundation of Yunnan University(No.ynuy201424)
文摘Activity is now playing a vital role in software processes. To ensure the high-level efficiency of software processes, a key point is to locate those activities that own bigger resource occupation probabilities with respect to average execution time, called delayed activities, and then improve them. To this end, we firstly propose an approach to locating delayed activities in software processes. Furthermore, we present a case study, which exhibits the high-level efficiency of the approach, to concretely illustrate this new solution. Some beneficial analysis and reasonable modification are developed in the end.
