Pub Date : 2010-11-01DOI: 10.1109/ISKE.2010.5680878
Ningsheng Gong, Zhigang Zhang
Generalization capability is a key flag to evaluate the performance of a learning system. Neural network ensemble can greatly improve the generalization capability of a learning system by training many neural networks and composing the result of them. In this paper, based on the theory of neural network ensemble, we present a constructive algorithm to improve the generalization capability of coverage-based neural networks. By construct positive-negative coverage group, the Generalization capability of the CBCNN-based networks can be greatly improved after constructed. Result of the theory analysis and experiments shows that our algorithm can greatly improve the generalization capability even when the initial classification capability of the neural networks is strong.
{"title":"Selective cover-based ensemble: Five maybe good enough","authors":"Ningsheng Gong, Zhigang Zhang","doi":"10.1109/ISKE.2010.5680878","DOIUrl":"https://doi.org/10.1109/ISKE.2010.5680878","url":null,"abstract":"Generalization capability is a key flag to evaluate the performance of a learning system. Neural network ensemble can greatly improve the generalization capability of a learning system by training many neural networks and composing the result of them. In this paper, based on the theory of neural network ensemble, we present a constructive algorithm to improve the generalization capability of coverage-based neural networks. By construct positive-negative coverage group, the Generalization capability of the CBCNN-based networks can be greatly improved after constructed. Result of the theory analysis and experiments shows that our algorithm can greatly improve the generalization capability even when the initial classification capability of the neural networks is strong.","PeriodicalId":6417,"journal":{"name":"2010 IEEE International Conference on Intelligent Systems and Knowledge Engineering","volume":"69 1","pages":"209-214"},"PeriodicalIF":0.0,"publicationDate":"2010-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73287256","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}
Pub Date : 2010-11-01DOI: 10.1109/ISKE.2010.5680794
Dao-jian Liu, Tian-min Huang
In this paper, based on the rotary iteration transformation in the simplex method, an artificial variable return-to-zero (RZ) algorithm is proposed to decide whether some condition constraint in linear programming problem is geometrically superfluous or not, as turns the decision problem of condition constraint geometric superfluity into such one whether condition constraint set of equations in its normalized form has such feasible basis solution in which every component corresponding to artificial variable is nought. Moreover, with the help of this method above, a feasible way to simplify linear programming problem is obtained.
{"title":"An artificial variable RZ decision method for geometric superfluity of condition constraint in linear programming problem","authors":"Dao-jian Liu, Tian-min Huang","doi":"10.1109/ISKE.2010.5680794","DOIUrl":"https://doi.org/10.1109/ISKE.2010.5680794","url":null,"abstract":"In this paper, based on the rotary iteration transformation in the simplex method, an artificial variable return-to-zero (RZ) algorithm is proposed to decide whether some condition constraint in linear programming problem is geometrically superfluous or not, as turns the decision problem of condition constraint geometric superfluity into such one whether condition constraint set of equations in its normalized form has such feasible basis solution in which every component corresponding to artificial variable is nought. Moreover, with the help of this method above, a feasible way to simplify linear programming problem is obtained.","PeriodicalId":6417,"journal":{"name":"2010 IEEE International Conference on Intelligent Systems and Knowledge Engineering","volume":"6 1","pages":"56-59"},"PeriodicalIF":0.0,"publicationDate":"2010-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"84988931","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}
Pub Date : 2010-11-01DOI: 10.1109/ISKE.2010.5680827
D. Wu, Jie Lu, Guangquan Zhang
Recommender systems aim to recommend items that are likely to be of interest to the user. In many business situations, complex items are described by hierarchical tree structures, which contain rich semantic information. To recommend hierarchical items accurately, the semantic information of the hierarchical tree structures must be considered comprehensively. In this study, a new hybrid recommendation approach for complex hierarchical tree structured items is proposed. In this approach, a comprehensive semantic similarity measure model for hierarchical tree structured items is developed. It is integrated with the traditional item-based collaborative filtering approach to generate recommendations.
{"title":"A hybrid recommendation approach for hierarchical items","authors":"D. Wu, Jie Lu, Guangquan Zhang","doi":"10.1109/ISKE.2010.5680827","DOIUrl":"https://doi.org/10.1109/ISKE.2010.5680827","url":null,"abstract":"Recommender systems aim to recommend items that are likely to be of interest to the user. In many business situations, complex items are described by hierarchical tree structures, which contain rich semantic information. To recommend hierarchical items accurately, the semantic information of the hierarchical tree structures must be considered comprehensively. In this study, a new hybrid recommendation approach for complex hierarchical tree structured items is proposed. In this approach, a comprehensive semantic similarity measure model for hierarchical tree structured items is developed. It is integrated with the traditional item-based collaborative filtering approach to generate recommendations.","PeriodicalId":6417,"journal":{"name":"2010 IEEE International Conference on Intelligent Systems and Knowledge Engineering","volume":"119 1","pages":"492-497"},"PeriodicalIF":0.0,"publicationDate":"2010-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"73405918","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}
Pub Date : 2010-11-01DOI: 10.1109/ISKE.2010.5680784
K. Vanhoof, B. Depaire
This paper provides a short review of various association rule classifiers (ARC) that have been developed over the past decade and the common structure behind most ARCs. Furthermore, different pruning and classification schemes used in various ARCs are reviewed and two ARCs are discussed which break with the standard structure behind ARC.
{"title":"Structure of association rule classifiers: a review","authors":"K. Vanhoof, B. Depaire","doi":"10.1109/ISKE.2010.5680784","DOIUrl":"https://doi.org/10.1109/ISKE.2010.5680784","url":null,"abstract":"This paper provides a short review of various association rule classifiers (ARC) that have been developed over the past decade and the common structure behind most ARCs. Furthermore, different pruning and classification schemes used in various ARCs are reviewed and two ARCs are discussed which break with the standard structure behind ARC.","PeriodicalId":6417,"journal":{"name":"2010 IEEE International Conference on Intelligent Systems and Knowledge Engineering","volume":"31 1","pages":"9-12"},"PeriodicalIF":0.0,"publicationDate":"2010-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76147255","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}
Pub Date : 2010-11-01DOI: 10.1109/ISKE.2010.5680812
Matilde Santos, V. López, F. Morata
The aim of this work is to describe an intelligent system based on fuzzy logic that is developed to control a quadrotor. A quadrotor is a helicopter with four rotors, that make the vehicle more stable but more complex to model and to control. The quadrotor has been used as a testing platform in the last years for various universities and research centres. A quadrotor has six degrees of freedom, three of them regarding the position: height, horizontal and vertical motions; and the other three are related to the orientation: pitch, roll and yaw. A fuzzy control is designed and implemented to control a simulation model of the quadrotor. The inputs are the desired values of the height, roll, pitch and yaw. The outputs are the power of each of the four rotors that is necessary to reach the specifications. Simulation results prove the efficiency of this intelligent control strategy.
{"title":"Intelligent fuzzy controller of a quadrotor","authors":"Matilde Santos, V. López, F. Morata","doi":"10.1109/ISKE.2010.5680812","DOIUrl":"https://doi.org/10.1109/ISKE.2010.5680812","url":null,"abstract":"The aim of this work is to describe an intelligent system based on fuzzy logic that is developed to control a quadrotor. A quadrotor is a helicopter with four rotors, that make the vehicle more stable but more complex to model and to control. The quadrotor has been used as a testing platform in the last years for various universities and research centres. A quadrotor has six degrees of freedom, three of them regarding the position: height, horizontal and vertical motions; and the other three are related to the orientation: pitch, roll and yaw. A fuzzy control is designed and implemented to control a simulation model of the quadrotor. The inputs are the desired values of the height, roll, pitch and yaw. The outputs are the power of each of the four rotors that is necessary to reach the specifications. Simulation results prove the efficiency of this intelligent control strategy.","PeriodicalId":6417,"journal":{"name":"2010 IEEE International Conference on Intelligent Systems and Knowledge Engineering","volume":"8 1","pages":"141-146"},"PeriodicalIF":0.0,"publicationDate":"2010-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79127366","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}
Pub Date : 2010-11-01DOI: 10.1109/ISKE.2010.5680814
José A. Sáez, J. Luengo, F. Herrera
The presence of noise is common in any real data set and may adversely affect the accuracy, construction time and complexity of the classifiers. Models built by Fuzzy Rule Based Classification Systems are recognised for their interpretability, but traditionally these methods have not considered the presence of noise in the data, so it would be interesting to quantify its effect on them. The aim of this contribution is to study the behavior and robustness of Fuzzy Rule Based Classification Systems in presence of noise. In order to do this, 69 synthetic data sets have been created from 23 data sets from the UCI repository. Different levels of noise have been introduced artificially in the class in order to analyze the FRBCSs when noise is present. The methods of Chi et al. and PDFC have been considered as a case study, analyzing the accuracy of the models created. From the results obtained, it is possible to deduce that Fuzzy Rule Based Classification Systems have a good tolerance to class noise.
{"title":"A first study on the noise impact in classes for Fuzzy Rule Based Classification Systems","authors":"José A. Sáez, J. Luengo, F. Herrera","doi":"10.1109/ISKE.2010.5680814","DOIUrl":"https://doi.org/10.1109/ISKE.2010.5680814","url":null,"abstract":"The presence of noise is common in any real data set and may adversely affect the accuracy, construction time and complexity of the classifiers. Models built by Fuzzy Rule Based Classification Systems are recognised for their interpretability, but traditionally these methods have not considered the presence of noise in the data, so it would be interesting to quantify its effect on them. The aim of this contribution is to study the behavior and robustness of Fuzzy Rule Based Classification Systems in presence of noise. In order to do this, 69 synthetic data sets have been created from 23 data sets from the UCI repository. Different levels of noise have been introduced artificially in the class in order to analyze the FRBCSs when noise is present. The methods of Chi et al. and PDFC have been considered as a case study, analyzing the accuracy of the models created. From the results obtained, it is possible to deduce that Fuzzy Rule Based Classification Systems have a good tolerance to class noise.","PeriodicalId":6417,"journal":{"name":"2010 IEEE International Conference on Intelligent Systems and Knowledge Engineering","volume":"16 1","pages":"153-158"},"PeriodicalIF":0.0,"publicationDate":"2010-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85642142","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}
Pub Date : 2010-11-01DOI: 10.1109/ISKE.2010.5680740
Yiming Jin, G. Mo, S. Zhang
High Dynamic Range Image (HDRI) is a type of images with wide luminosity levels in real scenes. It is very important to fuse the multiple photographs into a single HDRI. A method that constructs HDRI through recovering response curve of camera imaging system is proposed. The method can make effect of several important parameters to the synthesis results, and get a synthesis result for images missing the information of exposure time. As for the HDRI tone reproduction, an improved algorithm is presented in last part of the paper for processing HDRI and getting good performance
{"title":"Tone reproduction of High Dynamic Image","authors":"Yiming Jin, G. Mo, S. Zhang","doi":"10.1109/ISKE.2010.5680740","DOIUrl":"https://doi.org/10.1109/ISKE.2010.5680740","url":null,"abstract":"High Dynamic Range Image (HDRI) is a type of images with wide luminosity levels in real scenes. It is very important to fuse the multiple photographs into a single HDRI. A method that constructs HDRI through recovering response curve of camera imaging system is proposed. The method can make effect of several important parameters to the synthesis results, and get a synthesis result for images missing the information of exposure time. As for the HDRI tone reproduction, an improved algorithm is presented in last part of the paper for processing HDRI and getting good performance","PeriodicalId":6417,"journal":{"name":"2010 IEEE International Conference on Intelligent Systems and Knowledge Engineering","volume":"27 1","pages":"508-510"},"PeriodicalIF":0.0,"publicationDate":"2010-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"77820132","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}
Pub Date : 2010-11-01DOI: 10.1109/ISKE.2010.5680797
Junbo Zhang, Tianrui Li, Lei Wang
The rough set theory (RST) can deal with the approximations of an arbitrary subset of a universe by two definable subsets called, respectively, the lower and the upper approximations. Probabilistic rough sets (PRS) model is an extension of RST. In this paper, we propose a new matrix view of PRS and give the matrix representation of the lower and upper approximations of PRS. According to this matrix view, a new method is obtained for calculation of the lower and upper approximations under PRS. An example is given to illustrate processes of calculating the approximations of a concept based on the matrix point of view.
{"title":"A new method for calculation of the approximations under the probabilistic rough sets","authors":"Junbo Zhang, Tianrui Li, Lei Wang","doi":"10.1109/ISKE.2010.5680797","DOIUrl":"https://doi.org/10.1109/ISKE.2010.5680797","url":null,"abstract":"The rough set theory (RST) can deal with the approximations of an arbitrary subset of a universe by two definable subsets called, respectively, the lower and the upper approximations. Probabilistic rough sets (PRS) model is an extension of RST. In this paper, we propose a new matrix view of PRS and give the matrix representation of the lower and upper approximations of PRS. According to this matrix view, a new method is obtained for calculation of the lower and upper approximations under PRS. An example is given to illustrate processes of calculating the approximations of a concept based on the matrix point of view.","PeriodicalId":6417,"journal":{"name":"2010 IEEE International Conference on Intelligent Systems and Knowledge Engineering","volume":"11 6","pages":"73-76"},"PeriodicalIF":0.0,"publicationDate":"2010-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91398844","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}
Pub Date : 2010-11-01DOI: 10.1109/ISKE.2010.5680807
Peimin Wei, Jun Ye
In this paper, considered the information carried by the membership, the non-membership and the intuitionistic index in intuitionistic fuzzy sets (IFSs) as probability spaces, a modified cross-entropy measure between two IFSs is proposed to overcome the drawback of the intuitionistic fuzzy cross-entropy measure proposed by Vlachos and Sergiadis and the notion of symmetric discrimination information measure for IFSs is introduced. Finally, numerical examples are presented to demonstrate the efficiency of the proposed discrimination information measure for pattern recognition and medical diagnosis.
{"title":"Improved intuitionistic fuzzy cross-entropy and its application to pattern recognitions","authors":"Peimin Wei, Jun Ye","doi":"10.1109/ISKE.2010.5680807","DOIUrl":"https://doi.org/10.1109/ISKE.2010.5680807","url":null,"abstract":"In this paper, considered the information carried by the membership, the non-membership and the intuitionistic index in intuitionistic fuzzy sets (IFSs) as probability spaces, a modified cross-entropy measure between two IFSs is proposed to overcome the drawback of the intuitionistic fuzzy cross-entropy measure proposed by Vlachos and Sergiadis and the notion of symmetric discrimination information measure for IFSs is introduced. Finally, numerical examples are presented to demonstrate the efficiency of the proposed discrimination information measure for pattern recognition and medical diagnosis.","PeriodicalId":6417,"journal":{"name":"2010 IEEE International Conference on Intelligent Systems and Knowledge Engineering","volume":"9 1","pages":"114-116"},"PeriodicalIF":0.0,"publicationDate":"2010-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91144374","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}
Pub Date : 2010-11-01DOI: 10.1109/ISKE.2010.5680758
Zongyao Nie, Xiushui Ma, X. Qian, Junhai Jiang
This paper introduces the 110kV substations power supply system and devices of Anhui GITI Tire Company Ltd., analyses the principle of voltage management and proposes to improve the methods of voltage management quality for solving the difficulty of stable operation of power utilization for large chemical enterprises, which solution has application value.
{"title":"Voltage control of 110kV substation of enterprise","authors":"Zongyao Nie, Xiushui Ma, X. Qian, Junhai Jiang","doi":"10.1109/ISKE.2010.5680758","DOIUrl":"https://doi.org/10.1109/ISKE.2010.5680758","url":null,"abstract":"This paper introduces the 110kV substations power supply system and devices of Anhui GITI Tire Company Ltd., analyses the principle of voltage management and proposes to improve the methods of voltage management quality for solving the difficulty of stable operation of power utilization for large chemical enterprises, which solution has application value.","PeriodicalId":6417,"journal":{"name":"2010 IEEE International Conference on Intelligent Systems and Knowledge Engineering","volume":"110 1","pages":"588-591"},"PeriodicalIF":0.0,"publicationDate":"2010-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87696220","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：