Target maneuver recognition is a prerequisite for air combat situation awareness,trajectory prediction,threat assessment and maneuver decision.To get rid of the dependence of the current target maneuver recognition me...Target maneuver recognition is a prerequisite for air combat situation awareness,trajectory prediction,threat assessment and maneuver decision.To get rid of the dependence of the current target maneuver recognition method on empirical criteria and sample data,and automatically and adaptively complete the task of extracting the target maneuver pattern,in this paper,an air combat maneuver pattern extraction based on time series segmentation and clustering analysis is proposed by combining autoencoder,G-G clustering algorithm and the selective ensemble clustering analysis algorithm.Firstly,the autoencoder is used to extract key features of maneuvering trajectory to remove the impacts of redundant variables and reduce the data dimension;Then,taking the time information into account,the segmentation of Maneuver characteristic time series is realized with the improved FSTS-AEGG algorithm,and a large number of maneuver primitives are extracted;Finally,the maneuver primitives are grouped into some categories by using the selective ensemble multiple time series clustering algorithm,which can prove that each class represents a maneuver action.The maneuver pattern extraction method is applied to small scale air combat trajectory and can recognize and correctly partition at least 71.3%of maneuver actions,indicating that the method is effective and satisfies the requirements for engineering accuracy.In addition,this method can provide data support for various target maneuvering recognition methods proposed in the literature,greatly reduce the workload and improve the recognition accuracy.展开更多
High dimensional data clustering,with the inherent sparsity of data and the existence of noise,is a serious challenge for clustering algorithms.A new linear manifold clustering method was proposed to address this prob...High dimensional data clustering,with the inherent sparsity of data and the existence of noise,is a serious challenge for clustering algorithms.A new linear manifold clustering method was proposed to address this problem.The basic idea was to search the line manifold clusters hidden in datasets,and then fuse some of the line manifold clusters to construct higher dimensional manifold clusters.The orthogonal distance and the tangent distance were considered together as the linear manifold distance metrics. Spatial neighbor information was fully utilized to construct the original line manifold and optimize line manifolds during the line manifold cluster searching procedure.The results obtained from experiments over real and synthetic data sets demonstrate the superiority of the proposed method over some competing clustering methods in terms of accuracy and computation time.The proposed method is able to obtain high clustering accuracy for various data sets with different sizes,manifold dimensions and noise ratios,which confirms the anti-noise capability and high clustering accuracy of the proposed method for high dimensional data.展开更多
The Circle algorithm was proposed for large datasets.The idea of the algorithm is to find a set of vertices that are close to each other and far from other vertices.This algorithm makes use of the connection between c...The Circle algorithm was proposed for large datasets.The idea of the algorithm is to find a set of vertices that are close to each other and far from other vertices.This algorithm makes use of the connection between clustering aggregation and the problem of correlation clustering.The best deterministic approximation algorithm was provided for the variation of the correlation of clustering problem,and showed how sampling can be used to scale the algorithms for large datasets.An extensive empirical evaluation was given for the usefulness of the problem and the solutions.The results show that this method achieves more than 50% reduction in the running time without sacrificing the quality of the clustering.展开更多
Many classical clustering algorithms do good jobs on their prerequisite but do not scale well when being applied to deal with very large data sets(VLDS).In this work,a novel division and partition clustering method(DP...Many classical clustering algorithms do good jobs on their prerequisite but do not scale well when being applied to deal with very large data sets(VLDS).In this work,a novel division and partition clustering method(DP) was proposed to solve the problem.DP cut the source data set into data blocks,and extracted the eigenvector for each data block to form the local feature set.The local feature set was used in the second round of the characteristics polymerization process for the source data to find the global eigenvector.Ultimately according to the global eigenvector,the data set was assigned by criterion of minimum distance.The experimental results show that it is more robust than the conventional clusterings.Characteristics of not sensitive to data dimensions,distribution and number of nature clustering make it have a wide range of applications in clustering VLDS.展开更多
Energy-efficient data gathering in multi-hop wireless sensor networks was studied,considering that different node produces different amounts of data in realistic environments.A novel dominating set based clustering pr...Energy-efficient data gathering in multi-hop wireless sensor networks was studied,considering that different node produces different amounts of data in realistic environments.A novel dominating set based clustering protocol (DSCP) was proposed to solve the data gathering problem in this scenario.In DSCP,a node evaluates the potential lifetime of the network (from its local point of view) assuming that it acts as the cluster head,and claims to be a tentative cluster head if it maximizes the potential lifetime.When evaluating the potential lifetime of the network,a node considers not only its remaining energy,but also other factors including its traffic load,the number of its neighbors,and the traffic loads of its neighbors.A tentative cluster head becomes a final cluster head with a probability inversely proportional to the number of tentative cluster heads that cover its neighbors.The protocol can terminate in O(n/lg n) steps,and its total message complexity is O(n2/lg n).Simulation results show that DSCP can effectively prolong the lifetime of the network in multi-hop networks with unbalanced traffic load.Compared with EECT,the network lifetime is prolonged by 56.6% in average.展开更多
This paper introduces niching particle swarm optimiza- tion （nichePSO） into clustering analysis and puts forward a cluster- ing algorithm which uses nichePSO to optimize density functions. Firstly, this paper improv...This paper introduces niching particle swarm optimiza- tion （nichePSO） into clustering analysis and puts forward a cluster- ing algorithm which uses nichePSO to optimize density functions. Firstly, this paper improves main swarm training models and in- creases their ability of space searching. Secondly, the radius of sub-swarms is defined adaptively according to the actual clus- tering problem, which can be useful for the niches＇ forming and searching. At last, a novel method that distributes samples to the corresponding cluster is proposed. Numerical results illustrate that this algorithm based on the density function and nichePSO could cluster unbalanced density datasets into the correct clusters auto- matically and accurately.展开更多
DNS(domain name system) query log analysis has been a popular research topic in recent years. CLOPE, the represented transactional clustering algorithm, could be readily used for DNS query log mining. However, the alg...DNS(domain name system) query log analysis has been a popular research topic in recent years. CLOPE, the represented transactional clustering algorithm, could be readily used for DNS query log mining. However, the algorithm is inefficient when processing large scale data. The MR-CLOPE algorithm is proposed, which is an extension and improvement on CLOPE based on Map Reduce. Different from the previous parallel clustering method, a two-stage Map Reduce implementation framework is proposed. Each of the stage is implemented by one kind Map Reduce task. In the first stage, the DNS query logs are divided into multiple splits and the CLOPE algorithm is executed on each split. The second stage usually tends to iterate many times to merge the small clusters into bigger satisfactory ones. In these two stages, a novel partition process is designed to randomly spread out original sub clusters, which will be moved and merged in the map phrase of the second phase according to the defined merge criteria. In such way, the advantage of the original CLOPE algorithm is kept and its disadvantages are dealt with in the proposed framework to achieve more excellent clustering performance. The experiment results show that MR-CLOPE is not only faster but also has better clustering quality on DNS query logs compared with CLOPE.展开更多
Hierarchical clustering analysis based on statistic s is one of the most important mining algorithms, but the traditionary hierarchica l clustering method is based on global comparing, which only takes in Q clusteri n...Hierarchical clustering analysis based on statistic s is one of the most important mining algorithms, but the traditionary hierarchica l clustering method is based on global comparing, which only takes in Q clusteri ng while ignoring R clustering in practice, so it has some limitation especially when the number of sample and index is very large. Furthermore, because of igno ring the association between the different indexes, the clustering result is not good & true. In this paper, we present the model and the algorithm of two-level hierarchi cal clustering which integrates Q clustering with R clustering. Moreover, becaus e two-level hierarchical clustering is based on the respective clustering resul t of each class, the classification of the indexes directly effects on the a ccuracy of the final clustering result, how to appropriately classify the inde xes is the chief and difficult problem we must handle in advance. Although some literatures also have referred to the issue of the classificati on of the indexes, but the articles classify the indexes only according to their superficial signification, which is unscientific. The reasons are as follow s: First, the superficial signification of some indexes usually takes on different meanings and it is easy to be misapprehended by different person. Furthermore, t his classification method seldom make use of history data, the classification re sult is not so objective. Second, for some indexes, its superficial signification didn’t show any mean ings, so simply from the superficial signification, we can’t classify them to c ertain classes. Third, this classification method need the users have higher level knowledge of this field, otherwise it is difficult for the users to understand the signifi cation of some indexes, which sometimes is not available. So in this paper, to this question, we first use R clustering method to cluste ring indexes, dividing p dimension indexes into q classes, then adopt two-level clustering method to get the final result. Obviously, the classification result is more objective and accurate. Moreover, after the first step, we can get the relation of the different indexes and their interaction. We can also know under a certain class indexes, which samples can be clustering to a class. (These semi finished results sometimes are very useful.) The experiments also indicates the effective and accurate of the algorithms. And, the result of R clustering ca n be easily used for the later practice.展开更多
A new incremental clustering framework is presented, the basis of which is the induction as inverted deduction. Induction is inherently risky because it is not truth-preserving. If the clustering is considered as an i...A new incremental clustering framework is presented, the basis of which is the induction as inverted deduction. Induction is inherently risky because it is not truth-preserving. If the clustering is considered as an induction process, the key to build a valid clustering is to minimize the risk of clustering. From the viewpoint of modal logic, the clustering can be described as Kripke frames and Kripke models which are reflexive and symmetric. Based on the theory of modal logic, its properties can be described by system B in syntax. Thus, the risk of clustering can be calculated by the deduction relation of system B and proximity induction theorem described. Since the new proposed framework imposes no additional restrictive conditions of clustering algorithm, it is therefore a universal framework. An incremental clustering algorithm can be easily constructed by this framework from any given nonincremental clustering algorithm. The experiments show that the lower the a priori risk is, the more effective this framework is. It can be demonstrated that this framework is generally valid.展开更多
Outlier detection is an important task in data mining. In fact, it is difficult to find the clustering centers in some sophisticated multidimensional datasets and to measure the deviation degree of each potential outl...Outlier detection is an important task in data mining. In fact, it is difficult to find the clustering centers in some sophisticated multidimensional datasets and to measure the deviation degree of each potential outlier. In this work, an effective outlier detection method based on multi-dimensional clustering and local density(ODBMCLD) is proposed. ODBMCLD firstly identifies the center objects by the local density peak of data objects, and clusters the whole dataset based on the center objects. Then, outlier objects belonging to different clusters will be marked as candidates of abnormal data. Finally, the top N points among these abnormal candidates are chosen as final anomaly objects with high outlier factors. The feasibility and effectiveness of the method are verified by experiments.展开更多
To better understand different users' accessing intentions, a novel clustering and supervising method based on accessing path is presented. This method divides users' interest space to express the distribution...To better understand different users' accessing intentions, a novel clustering and supervising method based on accessing path is presented. This method divides users' interest space to express the distribution of users' interests, and directly to instruct the constructing process of web pages indexing for advanced performance.展开更多
In this paper,we introduce an adaptive clustering algorithm for intrusion detection based on wavecluster which was introduced by Gholamhosein in 1999 and used with success in image processing.Because of the non-statio...In this paper,we introduce an adaptive clustering algorithm for intrusion detection based on wavecluster which was introduced by Gholamhosein in 1999 and used with success in image processing.Because of the non-stationary characteristic of network traffic,we extend and develop an adaptive wavecluster algorithm for intrusion detection.Using the multiresolution property of wavelet transforms,we can effectively identify arbitrarily shaped clusters at different scales and degrees of detail,moreover,applying wavelet transform removes the noise from the original feature space and make more accurate cluster found.Experimental results on KDD-99 intrusion detection dataset show the efficiency and accuracy of this algorithm.A detection rate above 96% and a false alarm rate below 3% are achieved.展开更多
Data stream management system (DSMS) provides convenient solutions to the problem of processing continuous queries on data streams.Previous approaches for scheduling these queries and their operators assume that each ...Data stream management system (DSMS) provides convenient solutions to the problem of processing continuous queries on data streams.Previous approaches for scheduling these queries and their operators assume that each operator runs in separate thread or all operators combine in one query plan and run in a single thread.Both approaches suffer from severe drawbacks concerning the thread overhead and the stalls due to expensive operators.To overcome these drawbacks,a novel approach called clustered operators scheduling (COS) is proposed that adaptively clusters operators of the query plan into a number of groups based on their selectivity and computing cost using S-mean clustering.Experimental evaluation is provided to demonstrate the potential benefits of COS scheduling over the other scheduling strategies.COS can provide adaptive,flexible,reliable,scalable and robust design for continuous query processor.展开更多
Multi-laser-target tracking is an important subject in the field of signal processing of laser warners. A clustering method is applied to the measurement of laser warner, and the space-time fusion for measurements in ...Multi-laser-target tracking is an important subject in the field of signal processing of laser warners. A clustering method is applied to the measurement of laser warner, and the space-time fusion for measurements in the same cluster is accomplished. Real-time tracking of multi-laser-target and real-time picking of multi-laser-signal are introduced using data fusion of the measurements. A prototype device of the algorithm is built up. The results of experiments show that the algorithm is very effective.展开更多
Swarm robot systems are an important application of autonomous unmanned surface vehicles on water surfaces.For monitoring natural environments and conducting security activities within a certain range using a surface ...Swarm robot systems are an important application of autonomous unmanned surface vehicles on water surfaces.For monitoring natural environments and conducting security activities within a certain range using a surface vehicle,the swarm robot system is more efficient than the operation of a single object as the former can reduce cost and save time.It is necessary to detect adjacent surface obstacles robustly to operate a cluster of unmanned surface vehicles.For this purpose,a LiDAR(light detection and ranging)sensor is used as it can simultaneously obtain 3D information for all directions,relatively robustly and accurately,irrespective of the surrounding environmental conditions.Although the GPS(global-positioning-system)error range exists,obtaining measurements of the surface-vessel position can still ensure stability during platoon maneuvering.In this study,a three-layer convolutional neural network is applied to classify types of surface vehicles.The aim of this approach is to redefine the sparse 3D point cloud data as 2D image data with a connotative meaning and subsequently utilize this transformed data for object classification purposes.Hence,we have proposed a descriptor that converts the 3D point cloud data into 2D image data.To use this descriptor effectively,it is necessary to perform a clustering operation that separates the point clouds for each object.We developed voxel-based clustering for the point cloud clustering.Furthermore,using the descriptor,3D point cloud data can be converted into a 2D feature image,and the converted 2D image is provided as an input value to the network.We intend to verify the validity of the proposed 3D point cloud feature descriptor by using experimental data in the simulator.Furthermore,we explore the feasibility of real-time object classification within this framework.展开更多
An improved heuristic ant-clustering algorithm(HAC)is presented in this paper.A device of ’memory bank’ is proposed,which can bring forth heuristic knowledge guiding ant to move in the bi-dimension grid space.The de...An improved heuristic ant-clustering algorithm(HAC)is presented in this paper.A device of ’memory bank’ is proposed,which can bring forth heuristic knowledge guiding ant to move in the bi-dimension grid space.The device lowers the randomness of ants’ moving and avoids the producing of"un-assigned data object".We have made some experiments on real data sets and synthetic data sets.The results demonstrate that HAC has superiority in misclassification error rate and runtime over the classical algorithm.展开更多
基金supported by the National Natural Science Foundation of China (Project No.72301293)。
文摘Target maneuver recognition is a prerequisite for air combat situation awareness,trajectory prediction,threat assessment and maneuver decision.To get rid of the dependence of the current target maneuver recognition method on empirical criteria and sample data,and automatically and adaptively complete the task of extracting the target maneuver pattern,in this paper,an air combat maneuver pattern extraction based on time series segmentation and clustering analysis is proposed by combining autoencoder,G-G clustering algorithm and the selective ensemble clustering analysis algorithm.Firstly,the autoencoder is used to extract key features of maneuvering trajectory to remove the impacts of redundant variables and reduce the data dimension;Then,taking the time information into account,the segmentation of Maneuver characteristic time series is realized with the improved FSTS-AEGG algorithm,and a large number of maneuver primitives are extracted;Finally,the maneuver primitives are grouped into some categories by using the selective ensemble multiple time series clustering algorithm,which can prove that each class represents a maneuver action.The maneuver pattern extraction method is applied to small scale air combat trajectory and can recognize and correctly partition at least 71.3%of maneuver actions,indicating that the method is effective and satisfies the requirements for engineering accuracy.In addition,this method can provide data support for various target maneuvering recognition methods proposed in the literature,greatly reduce the workload and improve the recognition accuracy.
基金Project(60835005) supported by the National Nature Science Foundation of China
文摘High dimensional data clustering,with the inherent sparsity of data and the existence of noise,is a serious challenge for clustering algorithms.A new linear manifold clustering method was proposed to address this problem.The basic idea was to search the line manifold clusters hidden in datasets,and then fuse some of the line manifold clusters to construct higher dimensional manifold clusters.The orthogonal distance and the tangent distance were considered together as the linear manifold distance metrics. Spatial neighbor information was fully utilized to construct the original line manifold and optimize line manifolds during the line manifold cluster searching procedure.The results obtained from experiments over real and synthetic data sets demonstrate the superiority of the proposed method over some competing clustering methods in terms of accuracy and computation time.The proposed method is able to obtain high clustering accuracy for various data sets with different sizes,manifold dimensions and noise ratios,which confirms the anti-noise capability and high clustering accuracy of the proposed method for high dimensional data.
基金Projects(60873265,60903222) supported by the National Natural Science Foundation of China Project(IRT0661) supported by the Program for Changjiang Scholars and Innovative Research Team in University of China
文摘The Circle algorithm was proposed for large datasets.The idea of the algorithm is to find a set of vertices that are close to each other and far from other vertices.This algorithm makes use of the connection between clustering aggregation and the problem of correlation clustering.The best deterministic approximation algorithm was provided for the variation of the correlation of clustering problem,and showed how sampling can be used to scale the algorithms for large datasets.An extensive empirical evaluation was given for the usefulness of the problem and the solutions.The results show that this method achieves more than 50% reduction in the running time without sacrificing the quality of the clustering.
基金Projects(60903082,60975042)supported by the National Natural Science Foundation of ChinaProject(20070217043)supported by the Research Fund for the Doctoral Program of Higher Education of China
文摘Many classical clustering algorithms do good jobs on their prerequisite but do not scale well when being applied to deal with very large data sets(VLDS).In this work,a novel division and partition clustering method(DP) was proposed to solve the problem.DP cut the source data set into data blocks,and extracted the eigenvector for each data block to form the local feature set.The local feature set was used in the second round of the characteristics polymerization process for the source data to find the global eigenvector.Ultimately according to the global eigenvector,the data set was assigned by criterion of minimum distance.The experimental results show that it is more robust than the conventional clusterings.Characteristics of not sensitive to data dimensions,distribution and number of nature clustering make it have a wide range of applications in clustering VLDS.
基金Projects(61173169,61103203)supported by the National Natural Science Foundation of ChinaProject(NCET-10-0798)supported by the Program for New Century Excellent Talents in University of ChinaProject supported by the Post-doctoral Program and the Freedom Explore Program of Central South University,China
文摘Energy-efficient data gathering in multi-hop wireless sensor networks was studied,considering that different node produces different amounts of data in realistic environments.A novel dominating set based clustering protocol (DSCP) was proposed to solve the data gathering problem in this scenario.In DSCP,a node evaluates the potential lifetime of the network (from its local point of view) assuming that it acts as the cluster head,and claims to be a tentative cluster head if it maximizes the potential lifetime.When evaluating the potential lifetime of the network,a node considers not only its remaining energy,but also other factors including its traffic load,the number of its neighbors,and the traffic loads of its neighbors.A tentative cluster head becomes a final cluster head with a probability inversely proportional to the number of tentative cluster heads that cover its neighbors.The protocol can terminate in O(n/lg n) steps,and its total message complexity is O(n2/lg n).Simulation results show that DSCP can effectively prolong the lifetime of the network in multi-hop networks with unbalanced traffic load.Compared with EECT,the network lifetime is prolonged by 56.6% in average.
基金Supported by National Natural Science Foundation of China(60675039)National High Technology Research and Development Program of China(863 Program)(2006AA04Z217)Hundred Talents Program of Chinese Academy of Sciences
基金supported by the National Natural Science Foundation of China (708710157103100271171030)
文摘This paper introduces niching particle swarm optimiza- tion （nichePSO） into clustering analysis and puts forward a cluster- ing algorithm which uses nichePSO to optimize density functions. Firstly, this paper improves main swarm training models and in- creases their ability of space searching. Secondly, the radius of sub-swarms is defined adaptively according to the actual clus- tering problem, which can be useful for the niches＇ forming and searching. At last, a novel method that distributes samples to the corresponding cluster is proposed. Numerical results illustrate that this algorithm based on the density function and nichePSO could cluster unbalanced density datasets into the correct clusters auto- matically and accurately.
基金Project(61103046) supported in part by the National Natural Science Foundation of ChinaProject(B201312) supported by DHU Distinguished Young Professor Program,China+1 种基金Project(LY14F020007) supported by Zhejiang Provincial Natural Science Funds of ChinaProject(2014A610072) supported by the Natural Science Foundation of Ningbo City,China
文摘DNS(domain name system) query log analysis has been a popular research topic in recent years. CLOPE, the represented transactional clustering algorithm, could be readily used for DNS query log mining. However, the algorithm is inefficient when processing large scale data. The MR-CLOPE algorithm is proposed, which is an extension and improvement on CLOPE based on Map Reduce. Different from the previous parallel clustering method, a two-stage Map Reduce implementation framework is proposed. Each of the stage is implemented by one kind Map Reduce task. In the first stage, the DNS query logs are divided into multiple splits and the CLOPE algorithm is executed on each split. The second stage usually tends to iterate many times to merge the small clusters into bigger satisfactory ones. In these two stages, a novel partition process is designed to randomly spread out original sub clusters, which will be moved and merged in the map phrase of the second phase according to the defined merge criteria. In such way, the advantage of the original CLOPE algorithm is kept and its disadvantages are dealt with in the proposed framework to achieve more excellent clustering performance. The experiment results show that MR-CLOPE is not only faster but also has better clustering quality on DNS query logs compared with CLOPE.
文摘Hierarchical clustering analysis based on statistic s is one of the most important mining algorithms, but the traditionary hierarchica l clustering method is based on global comparing, which only takes in Q clusteri ng while ignoring R clustering in practice, so it has some limitation especially when the number of sample and index is very large. Furthermore, because of igno ring the association between the different indexes, the clustering result is not good & true. In this paper, we present the model and the algorithm of two-level hierarchi cal clustering which integrates Q clustering with R clustering. Moreover, becaus e two-level hierarchical clustering is based on the respective clustering resul t of each class, the classification of the indexes directly effects on the a ccuracy of the final clustering result, how to appropriately classify the inde xes is the chief and difficult problem we must handle in advance. Although some literatures also have referred to the issue of the classificati on of the indexes, but the articles classify the indexes only according to their superficial signification, which is unscientific. The reasons are as follow s: First, the superficial signification of some indexes usually takes on different meanings and it is easy to be misapprehended by different person. Furthermore, t his classification method seldom make use of history data, the classification re sult is not so objective. Second, for some indexes, its superficial signification didn’t show any mean ings, so simply from the superficial signification, we can’t classify them to c ertain classes. Third, this classification method need the users have higher level knowledge of this field, otherwise it is difficult for the users to understand the signifi cation of some indexes, which sometimes is not available. So in this paper, to this question, we first use R clustering method to cluste ring indexes, dividing p dimension indexes into q classes, then adopt two-level clustering method to get the final result. Obviously, the classification result is more objective and accurate. Moreover, after the first step, we can get the relation of the different indexes and their interaction. We can also know under a certain class indexes, which samples can be clustering to a class. (These semi finished results sometimes are very useful.) The experiments also indicates the effective and accurate of the algorithms. And, the result of R clustering ca n be easily used for the later practice.
基金supported by the National High-Tech Research and Development Program of China(2006AA12A106).
文摘A new incremental clustering framework is presented, the basis of which is the induction as inverted deduction. Induction is inherently risky because it is not truth-preserving. If the clustering is considered as an induction process, the key to build a valid clustering is to minimize the risk of clustering. From the viewpoint of modal logic, the clustering can be described as Kripke frames and Kripke models which are reflexive and symmetric. Based on the theory of modal logic, its properties can be described by system B in syntax. Thus, the risk of clustering can be calculated by the deduction relation of system B and proximity induction theorem described. Since the new proposed framework imposes no additional restrictive conditions of clustering algorithm, it is therefore a universal framework. An incremental clustering algorithm can be easily constructed by this framework from any given nonincremental clustering algorithm. The experiments show that the lower the a priori risk is, the more effective this framework is. It can be demonstrated that this framework is generally valid.
基金Project(61362021)supported by the National Natural Science Foundation of ChinaProject(2016GXNSFAA380149)supported by Natural Science Foundation of Guangxi Province,China+1 种基金Projects(2016YJCXB02,2017YJCX34)supported by Innovation Project of GUET Graduate Education,ChinaProject(2011KF11)supported by the Key Laboratory of Cognitive Radio and Information Processing,Ministry of Education,China
文摘Outlier detection is an important task in data mining. In fact, it is difficult to find the clustering centers in some sophisticated multidimensional datasets and to measure the deviation degree of each potential outlier. In this work, an effective outlier detection method based on multi-dimensional clustering and local density(ODBMCLD) is proposed. ODBMCLD firstly identifies the center objects by the local density peak of data objects, and clusters the whole dataset based on the center objects. Then, outlier objects belonging to different clusters will be marked as candidates of abnormal data. Finally, the top N points among these abnormal candidates are chosen as final anomaly objects with high outlier factors. The feasibility and effectiveness of the method are verified by experiments.
文摘To better understand different users' accessing intentions, a novel clustering and supervising method based on accessing path is presented. This method divides users' interest space to express the distribution of users' interests, and directly to instruct the constructing process of web pages indexing for advanced performance.
文摘In this paper,we introduce an adaptive clustering algorithm for intrusion detection based on wavecluster which was introduced by Gholamhosein in 1999 and used with success in image processing.Because of the non-stationary characteristic of network traffic,we extend and develop an adaptive wavecluster algorithm for intrusion detection.Using the multiresolution property of wavelet transforms,we can effectively identify arbitrarily shaped clusters at different scales and degrees of detail,moreover,applying wavelet transform removes the noise from the original feature space and make more accurate cluster found.Experimental results on KDD-99 intrusion detection dataset show the efficiency and accuracy of this algorithm.A detection rate above 96% and a false alarm rate below 3% are achieved.
基金Project(50275150) supported by the National Natural Science Foundation of ChinaProject(20040533035) supported by the National Research Foundation for the Doctoral Program of Higher Education of China
文摘Data stream management system (DSMS) provides convenient solutions to the problem of processing continuous queries on data streams.Previous approaches for scheduling these queries and their operators assume that each operator runs in separate thread or all operators combine in one query plan and run in a single thread.Both approaches suffer from severe drawbacks concerning the thread overhead and the stalls due to expensive operators.To overcome these drawbacks,a novel approach called clustered operators scheduling (COS) is proposed that adaptively clusters operators of the query plan into a number of groups based on their selectivity and computing cost using S-mean clustering.Experimental evaluation is provided to demonstrate the potential benefits of COS scheduling over the other scheduling strategies.COS can provide adaptive,flexible,reliable,scalable and robust design for continuous query processor.
基金University Doctor Subject Foundation of China (20060699024)
文摘Multi-laser-target tracking is an important subject in the field of signal processing of laser warners. A clustering method is applied to the measurement of laser warner, and the space-time fusion for measurements in the same cluster is accomplished. Real-time tracking of multi-laser-target and real-time picking of multi-laser-signal are introduced using data fusion of the measurements. A prototype device of the algorithm is built up. The results of experiments show that the algorithm is very effective.
基金supported by the Future Challenge Program through the Agency for Defense Development funded by the Defense Acquisition Program Administration (No.UC200015RD)。
文摘Swarm robot systems are an important application of autonomous unmanned surface vehicles on water surfaces.For monitoring natural environments and conducting security activities within a certain range using a surface vehicle,the swarm robot system is more efficient than the operation of a single object as the former can reduce cost and save time.It is necessary to detect adjacent surface obstacles robustly to operate a cluster of unmanned surface vehicles.For this purpose,a LiDAR(light detection and ranging)sensor is used as it can simultaneously obtain 3D information for all directions,relatively robustly and accurately,irrespective of the surrounding environmental conditions.Although the GPS(global-positioning-system)error range exists,obtaining measurements of the surface-vessel position can still ensure stability during platoon maneuvering.In this study,a three-layer convolutional neural network is applied to classify types of surface vehicles.The aim of this approach is to redefine the sparse 3D point cloud data as 2D image data with a connotative meaning and subsequently utilize this transformed data for object classification purposes.Hence,we have proposed a descriptor that converts the 3D point cloud data into 2D image data.To use this descriptor effectively,it is necessary to perform a clustering operation that separates the point clouds for each object.We developed voxel-based clustering for the point cloud clustering.Furthermore,using the descriptor,3D point cloud data can be converted into a 2D feature image,and the converted 2D image is provided as an input value to the network.We intend to verify the validity of the proposed 3D point cloud feature descriptor by using experimental data in the simulator.Furthermore,we explore the feasibility of real-time object classification within this framework.
文摘An improved heuristic ant-clustering algorithm(HAC)is presented in this paper.A device of ’memory bank’ is proposed,which can bring forth heuristic knowledge guiding ant to move in the bi-dimension grid space.The device lowers the randomness of ants’ moving and avoids the producing of"un-assigned data object".We have made some experiments on real data sets and synthetic data sets.The results demonstrate that HAC has superiority in misclassification error rate and runtime over the classical algorithm.
