We propose a novel scheme, called Cooperative Semantic Caching Scheme (CSCS) for multi-level super-peer networks that support hierarchical P2P architecture. Peers share information according the semantic proximity between peers and between shared files to self-organise into clustered groups. In this paper we tackle the problem of exploiting the semantic locality of peer requests to improve the performance of a P2P network by the use of cooperative semantic caches. Such caches group together peers with similar interests as well as files with similar request patterns. Simulation experiments show that the CSCS mechanism achieves significant improvements in terms of access latency and global cache hit ratio.
{"title":"A Cooperative Semantic Caching Scheme for Super-Peer Networks","authors":"Yunsong Tan, Y. Wu","doi":"10.1109/CINC.2009.233","DOIUrl":"https://doi.org/10.1109/CINC.2009.233","url":null,"abstract":"We propose a novel scheme, called Cooperative Semantic Caching Scheme (CSCS) for multi-level super-peer networks that support hierarchical P2P architecture. Peers share information according the semantic proximity between peers and between shared files to self-organise into clustered groups. In this paper we tackle the problem of exploiting the semantic locality of peer requests to improve the performance of a P2P network by the use of cooperative semantic caches. Such caches group together peers with similar interests as well as files with similar request patterns. Simulation experiments show that the CSCS mechanism achieves significant improvements in terms of access latency and global cache hit ratio.","PeriodicalId":173506,"journal":{"name":"2009 International Conference on Computational Intelligence and Natural Computing","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131807054","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Overfitting is an important topic in Neural Network. Internal Symmetry Networks are a new modern Cellular Neural Networks inspired by the phenomenon of internal symmetry in quantum physics. Recurrent Internal Symmetry Networks are just studied very recently. In this paper, overfitting in recurrent cycles of Internal Symmetry Networks is analyzed. Back propagation is trained for an image processing task.
{"title":"Problem and Strategy: Overfitting in Recurrent Cycles of Internal Symmetry Networks by Back Propagation","authors":"Guanzhong Li","doi":"10.1109/CINC.2009.258","DOIUrl":"https://doi.org/10.1109/CINC.2009.258","url":null,"abstract":"Overfitting is an important topic in Neural Network. Internal Symmetry Networks are a new modern Cellular Neural Networks inspired by the phenomenon of internal symmetry in quantum physics. Recurrent Internal Symmetry Networks are just studied very recently. In this paper, overfitting in recurrent cycles of Internal Symmetry Networks is analyzed. Back propagation is trained for an image processing task.","PeriodicalId":173506,"journal":{"name":"2009 International Conference on Computational Intelligence and Natural Computing","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133380777","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Face recognition has become one of the latest research subjects of pattern recognition and image processing. Although many face recognition techniques have been proposed and many achievements have been obtained, we can’t get high recognition rate due to the changes of face expression, location, direction and light. In this paper we study human face recognition based on ensemble techniques. In order to improve diversity of component classifiers, the idea of bit-plane decomposition is used and moving window classifier is used as a basic individual classifier. The quantized pattern representations’ layers are used jointly to make a decision. And we mainly study several fused methods which include product, sum, majority vote, max, min and median rules. Experiments results with face images databases show that fusion of multiple classifiers has good classification performance. Moreover, we compare different multiple classifier schemes with other human face recognition methods.
{"title":"Ensemble Methods of Face Recognition Based on Bit-plane Decomposition","authors":"Kai Li, Lingxiao Wang","doi":"10.1109/CINC.2009.216","DOIUrl":"https://doi.org/10.1109/CINC.2009.216","url":null,"abstract":"Face recognition has become one of the latest research subjects of pattern recognition and image processing. Although many face recognition techniques have been proposed and many achievements have been obtained, we can’t get high recognition rate due to the changes of face expression, location, direction and light. In this paper we study human face recognition based on ensemble techniques. In order to improve diversity of component classifiers, the idea of bit-plane decomposition is used and moving window classifier is used as a basic individual classifier. The quantized pattern representations’ layers are used jointly to make a decision. And we mainly study several fused methods which include product, sum, majority vote, max, min and median rules. Experiments results with face images databases show that fusion of multiple classifiers has good classification performance. Moreover, we compare different multiple classifier schemes with other human face recognition methods.","PeriodicalId":173506,"journal":{"name":"2009 International Conference on Computational Intelligence and Natural Computing","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131797020","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In order to balance loadings in heterogenous parallel processing systems, a new task scheduling algorithm, weighted least connection genetic algorithm (WLGA), is proposed. WLGA algorithm uses the genetic algorithm to improve the weighted least connection algorithm (WLCA), it overcomes deficiencies of WLCA algorithms and provides functions of dynamic control to schedule tasks so that the distribution problem of N processors is solved effectively. The experimental result shows the improved algorithm WLGA is superior to basic genetic algorithm and WLCA algorithm.
{"title":"An Efficient Dynamic Load Balancing Scheme for Heterogenous Processing System","authors":"Xiaonian Tong, Wanneng Shu","doi":"10.1109/CINC.2009.77","DOIUrl":"https://doi.org/10.1109/CINC.2009.77","url":null,"abstract":"In order to balance loadings in heterogenous parallel processing systems, a new task scheduling algorithm, weighted least connection genetic algorithm (WLGA), is proposed. WLGA algorithm uses the genetic algorithm to improve the weighted least connection algorithm (WLCA), it overcomes deficiencies of WLCA algorithms and provides functions of dynamic control to schedule tasks so that the distribution problem of N processors is solved effectively. The experimental result shows the improved algorithm WLGA is superior to basic genetic algorithm and WLCA algorithm.","PeriodicalId":173506,"journal":{"name":"2009 International Conference on Computational Intelligence and Natural Computing","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132430438","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Current classification methods are based on the “Bag of Words” (BOW) representation, which only accounts for term frequency in the documents, and ignores important semantic relationships between key terms. In this paper, we proposed a system that uses ontologies and Natural Language Processing techniques to index texts. Traditional BOW matrix is replaced by “Bag of Concepts” (BOC). For this purpose, we developed fully automated methods for mapping keywords to their corresponding ontology concepts. Support Vector Machine a successful machine learning technique is used for classification. Experimental results shows that our proposed method dose improve text classification performance significantly
{"title":"Applying RDF Ontologies to Improve Text Classification","authors":"Wang Xiaoyue, Bai Rujiang","doi":"10.1109/CINC.2009.115","DOIUrl":"https://doi.org/10.1109/CINC.2009.115","url":null,"abstract":"Current classification methods are based on the “Bag of Words” (BOW) representation, which only accounts for term frequency in the documents, and ignores important semantic relationships between key terms. In this paper, we proposed a system that uses ontologies and Natural Language Processing techniques to index texts. Traditional BOW matrix is replaced by “Bag of Concepts” (BOC). For this purpose, we developed fully automated methods for mapping keywords to their corresponding ontology concepts. Support Vector Machine a successful machine learning technique is used for classification. Experimental results shows that our proposed method dose improve text classification performance significantly","PeriodicalId":173506,"journal":{"name":"2009 International Conference on Computational Intelligence and Natural Computing","volume":"260 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131536927","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper describes the optimization solution to improve the dynamic stiffness of the deformable mirror supporting structure. With the methods of Finite element analysis(FEA), Orthogonal experiment and BP Neural Network, the relationship between the structure parameters of the deformable mirror supporting structure and its resonate frequency is built. With this relationship and Genetic Algorithm(GA) optimal design, a group of reasonable structure parameters are found that can improve the dynamic stiffness of the deformable mirror supporting structure.
{"title":"Optimal Design for the Supporting Structure of the Deformable Mirror","authors":"Fu Zhao, Y. Gong, Li Zhang, H. Xiang, Ping Wang","doi":"10.1109/CINC.2009.207","DOIUrl":"https://doi.org/10.1109/CINC.2009.207","url":null,"abstract":"This paper describes the optimization solution to improve the dynamic stiffness of the deformable mirror supporting structure. With the methods of Finite element analysis(FEA), Orthogonal experiment and BP Neural Network, the relationship between the structure parameters of the deformable mirror supporting structure and its resonate frequency is built. With this relationship and Genetic Algorithm(GA) optimal design, a group of reasonable structure parameters are found that can improve the dynamic stiffness of the deformable mirror supporting structure.","PeriodicalId":173506,"journal":{"name":"2009 International Conference on Computational Intelligence and Natural Computing","volume":"108 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131545106","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
License plate location is one of the key link in the license plate recognition process. Whether the plate location is successful and how accurate is the location decide directly the recognition and the effects in the latter part. For the license plate area has a high density difference in the difference image,we have put forward an algorithm for the license plate location based on the density and projection. The results show that this method can quickly and correctly locate the license plate area.
{"title":"Car License Plate Location Based on the Density and Projection","authors":"J. Su, Zheng Ma","doi":"10.1109/CINC.2009.70","DOIUrl":"https://doi.org/10.1109/CINC.2009.70","url":null,"abstract":"License plate location is one of the key link in the license plate recognition process. Whether the plate location is successful and how accurate is the location decide directly the recognition and the effects in the latter part. For the license plate area has a high density difference in the difference image,we have put forward an algorithm for the license plate location based on the density and projection. The results show that this method can quickly and correctly locate the license plate area.","PeriodicalId":173506,"journal":{"name":"2009 International Conference on Computational Intelligence and Natural Computing","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133005258","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Chinese wines can be classification or graded by the micrographs. Micrographs of Chinese wines show floccule,stick and granule of variant shape and size. Different wines have variant microstructure and micrographs, we study the classification of Chinese wines based on the micrographs. Shape and structure of wines’ particles in microstructure is the most important feature for recognition and classification of wines. So we introduce a new feature extraction method which can describe the structure and region shape of micrograph efficiently. First, the micrographs are enhanced using total variation denoising, and segmented using relative entropy thresholding. Then features are extracted using proposed method in the paper based on area, perimeter and traditional shape feature. Eight kinds total 26 features are selected. Finally, Chinese wine classification system based on micrograph using combination of shape and structure features and BP neural network have been presented. We compare the recognition results for different choices of features (traditional shape features or proposed features). The experimental results show that the better classification rate have been achieved using the combinational features proposed in this paper.
{"title":"Shape and Structure Features Based Chinese Wine Classification","authors":"Yi Wan, Xingbo Sun, Rong Guo","doi":"10.1109/CINC.2009.191","DOIUrl":"https://doi.org/10.1109/CINC.2009.191","url":null,"abstract":"Chinese wines can be classification or graded by the micrographs. Micrographs of Chinese wines show floccule,stick and granule of variant shape and size. Different wines have variant microstructure and micrographs, we study the classification of Chinese wines based on the micrographs. Shape and structure of wines’ particles in microstructure is the most important feature for recognition and classification of wines. So we introduce a new feature extraction method which can describe the structure and region shape of micrograph efficiently. First, the micrographs are enhanced using total variation denoising, and segmented using relative entropy thresholding. Then features are extracted using proposed method in the paper based on area, perimeter and traditional shape feature. Eight kinds total 26 features are selected. Finally, Chinese wine classification system based on micrograph using combination of shape and structure features and BP neural network have been presented. We compare the recognition results for different choices of features (traditional shape features or proposed features). The experimental results show that the better classification rate have been achieved using the combinational features proposed in this paper.","PeriodicalId":173506,"journal":{"name":"2009 International Conference on Computational Intelligence and Natural Computing","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132800434","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
With the rapid development of the Internet, P2P technology has changed the traditional concept of Internet; this technology has become the dominant flow and greatly satisfied the needs of the users. At the same time, P2P worm has become one of the serious threats. This article include characteristics of P2P worms, classification, propagation model, detection and defense mechanism, also analyze some advantages and disadvantages of simulation software. At last this article discusses the direction for future research.
{"title":"The Research of Worms in P2P Networks","authors":"Xie Chunyan, Yin Zhiyu","doi":"10.1109/CINC.2009.248","DOIUrl":"https://doi.org/10.1109/CINC.2009.248","url":null,"abstract":"With the rapid development of the Internet, P2P technology has changed the traditional concept of Internet; this technology has become the dominant flow and greatly satisfied the needs of the users. At the same time, P2P worm has become one of the serious threats. This article include characteristics of P2P worms, classification, propagation model, detection and defense mechanism, also analyze some advantages and disadvantages of simulation software. At last this article discusses the direction for future research.","PeriodicalId":173506,"journal":{"name":"2009 International Conference on Computational Intelligence and Natural Computing","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121472275","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper presents a method of using nonlinear decision function to improve the performance of AdaBoost with SVM based weak learners. Compared with the existing AdaBoostSVM methods,this method, named ERBF-AdaBoostSVM, has advantages of higher hate rate and better generalization performance. This method also provides non-linear separator in the weak learner space and classifies accurately more examples. Experimental results demonstrated that ERBF-AdaBoostSVM achieve better generalization performance and higher hate rate than the existing SVM and AdaBoostSVM methods.
{"title":"An AdaBoost Algorithm with SVM Based on Nonlinear Decision Function","authors":"W. Wu, Z. Yanan, Wu Linlin","doi":"10.1109/CINC.2009.256","DOIUrl":"https://doi.org/10.1109/CINC.2009.256","url":null,"abstract":"This paper presents a method of using nonlinear decision function to improve the performance of AdaBoost with SVM based weak learners. Compared with the existing AdaBoostSVM methods,this method, named ERBF-AdaBoostSVM, has advantages of higher hate rate and better generalization performance. This method also provides non-linear separator in the weak learner space and classifies accurately more examples. Experimental results demonstrated that ERBF-AdaBoostSVM achieve better generalization performance and higher hate rate than the existing SVM and AdaBoostSVM methods.","PeriodicalId":173506,"journal":{"name":"2009 International Conference on Computational Intelligence and Natural Computing","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-06-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117037041","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：