Pub Date : 2010-07-11DOI: 10.1109/ICMLC.2010.5580972
Zhiting Guo, Hui Wang, Zhiwen Lin, Xiaoxia Guo
Counting all common subsequences (ACS) was proposed as a similarity measurement, which is conceptually different from the sequence kernel (SK) in that ACS only considers the occurrence of subsequences while SK uses the frequency of occurrences of subsequences. This difference evidently results in significant performance variety. ACS has been very competitive in the kNN classifier, however, its performance with kernel machine has been rarely investigated. This is due to the fact that whether ACS is suitable for a kernel classifier is not clear. To this end, this paper firstly proves that ACS is a valid kernel, with a delicate analysis. Then, ACS is further proved to be a good kernel with a comparison with SK in the support vector machine.
{"title":"A study of all common subsequences in kernel machine","authors":"Zhiting Guo, Hui Wang, Zhiwen Lin, Xiaoxia Guo","doi":"10.1109/ICMLC.2010.5580972","DOIUrl":"https://doi.org/10.1109/ICMLC.2010.5580972","url":null,"abstract":"Counting all common subsequences (ACS) was proposed as a similarity measurement, which is conceptually different from the sequence kernel (SK) in that ACS only considers the occurrence of subsequences while SK uses the frequency of occurrences of subsequences. This difference evidently results in significant performance variety. ACS has been very competitive in the kNN classifier, however, its performance with kernel machine has been rarely investigated. This is due to the fact that whether ACS is suitable for a kernel classifier is not clear. To this end, this paper firstly proves that ACS is a valid kernel, with a delicate analysis. Then, ACS is further proved to be a good kernel with a comparison with SK in the support vector machine.","PeriodicalId":126080,"journal":{"name":"2010 International Conference on Machine Learning and Cybernetics","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129271949","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 : 2010-07-11DOI: 10.1109/ICMLC.2010.5580917
Yuxiu Yan, Ming Wang, Zimin Jin, Xiu-Juan Hu, Jianwei Tao
According to wind resistance boundary-layer theory and form resistance, we designed the three-dimensional seamless jersey which has the function of resistant of wind resistance. Through the experiment of the strength measurement by the hydromechanics wind tunnel, we researched the three-dimensional seamless jersey' resistant of wind resistance. The experiment result indicated: the toward structure of jersey's sleeve cuff satisfied the body feature of the racer in the riding states and reduce the wind resistance effectively. The seamless jersey knitted by seamless knitting machine also has the function of reducing the wind resistance.
{"title":"The research of three-dimensional seamless jersey's resistant of wind resistance","authors":"Yuxiu Yan, Ming Wang, Zimin Jin, Xiu-Juan Hu, Jianwei Tao","doi":"10.1109/ICMLC.2010.5580917","DOIUrl":"https://doi.org/10.1109/ICMLC.2010.5580917","url":null,"abstract":"According to wind resistance boundary-layer theory and form resistance, we designed the three-dimensional seamless jersey which has the function of resistant of wind resistance. Through the experiment of the strength measurement by the hydromechanics wind tunnel, we researched the three-dimensional seamless jersey' resistant of wind resistance. The experiment result indicated: the toward structure of jersey's sleeve cuff satisfied the body feature of the racer in the riding states and reduce the wind resistance effectively. The seamless jersey knitted by seamless knitting machine also has the function of reducing the wind resistance.","PeriodicalId":126080,"journal":{"name":"2010 International Conference on Machine Learning and Cybernetics","volume":"102 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124744574","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}
Statistical learning theory on probability space is an important part of Machine Learning. Based on the key theorem, the bounds of uniform convergence have significant meaning. These bounds determine generalization ability of the learning machines utilizing the empirical risk minimization induction principle. In this paper, the bounds on the risk for real-valued loss function of the learning processes on possibility space are discussed, and the rate of uniform convergence is estimated.
{"title":"The bounds on the risk for real-valued loss functions on possibility space","authors":"Peng Wang, Yun-Chao Bai, Chun-Qin Zhang, Cai-Li Zhou","doi":"10.1109/ICMLC.2010.5580968","DOIUrl":"https://doi.org/10.1109/ICMLC.2010.5580968","url":null,"abstract":"Statistical learning theory on probability space is an important part of Machine Learning. Based on the key theorem, the bounds of uniform convergence have significant meaning. These bounds determine generalization ability of the learning machines utilizing the empirical risk minimization induction principle. In this paper, the bounds on the risk for real-valued loss function of the learning processes on possibility space are discussed, and the rate of uniform convergence is estimated.","PeriodicalId":126080,"journal":{"name":"2010 International Conference on Machine Learning and Cybernetics","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129886847","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 : 2010-07-11DOI: 10.1109/ICMLC.2010.5580791
Chi-Kuang Hwang, K. Huang, Kuo-Bin Lin, Bore-Kuen Lee
For the continuous infinite horizon time-invariant linear quadratic regulator problem (LQR), in the paper, the optimal state feedback controller based on the estimated state of the observer can be decoupled by the proposed approach which resulting one continuous time algebraic Riccati equation (CARE) for the controller design and one matrix equality equation (MEE) for the observer design. A coupling term related the CARE of the controller is found to be existed in the MEE of the observer. Unlike the separate principle to design the controller and observer separately without any coupling term, the design of the observer should consider the coupling term related to the CARE of the controller. The coupling problem between the controller and the observer usually exists in the linear matrix inequality (LMI) approach, and it is the main problem to be solved. The two-stage scheme is popular in the LMI approach, and the proposed method is similar to it, but adopting equality instead of inequality.
{"title":"Observer base linear quadratic regulation with estimated state feedback control","authors":"Chi-Kuang Hwang, K. Huang, Kuo-Bin Lin, Bore-Kuen Lee","doi":"10.1109/ICMLC.2010.5580791","DOIUrl":"https://doi.org/10.1109/ICMLC.2010.5580791","url":null,"abstract":"For the continuous infinite horizon time-invariant linear quadratic regulator problem (LQR), in the paper, the optimal state feedback controller based on the estimated state of the observer can be decoupled by the proposed approach which resulting one continuous time algebraic Riccati equation (CARE) for the controller design and one matrix equality equation (MEE) for the observer design. A coupling term related the CARE of the controller is found to be existed in the MEE of the observer. Unlike the separate principle to design the controller and observer separately without any coupling term, the design of the observer should consider the coupling term related to the CARE of the controller. The coupling problem between the controller and the observer usually exists in the linear matrix inequality (LMI) approach, and it is the main problem to be solved. The two-stage scheme is popular in the LMI approach, and the proposed method is similar to it, but adopting equality instead of inequality.","PeriodicalId":126080,"journal":{"name":"2010 International Conference on Machine Learning and Cybernetics","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130333735","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 : 2010-07-11DOI: 10.1109/ICMLC.2010.5580474
C. Fang, Bo Wu, Jung-Ming Wang, Sei-Wang Chen
This paper presents a system for predicting dangerous driving events while driving on an expressway. There are three major tasks involved in the prediction system: (1) how to perceive driving events on the input sequence of driving conditions, (2) how to represent driving events, and (3) how to interpret driving events to decide whether or not they are hazardous. A directed acyclic graph, called the attributed driving relational map (ADRM), is introduced to represent driving events. The ADRM chronicles a driving event in terms of driving conditions. The prediction system evaluates the driving event to determine whether it is perilous or not by matching its ADRM against those of known dangerous driving events preserved in a database using a fuzzy attributed map matching technique. The database can automatically augment by including new dangerous driving events that approved any of the predefined danger criteria. A series of experiments with synthetic examples generated by a driving simulator have been conducted to demonstrate the feasibility and rationality of the proposed system.
{"title":"Dangerous driving event prediction on expressways using fuzzy attributed map matching","authors":"C. Fang, Bo Wu, Jung-Ming Wang, Sei-Wang Chen","doi":"10.1109/ICMLC.2010.5580474","DOIUrl":"https://doi.org/10.1109/ICMLC.2010.5580474","url":null,"abstract":"This paper presents a system for predicting dangerous driving events while driving on an expressway. There are three major tasks involved in the prediction system: (1) how to perceive driving events on the input sequence of driving conditions, (2) how to represent driving events, and (3) how to interpret driving events to decide whether or not they are hazardous. A directed acyclic graph, called the attributed driving relational map (ADRM), is introduced to represent driving events. The ADRM chronicles a driving event in terms of driving conditions. The prediction system evaluates the driving event to determine whether it is perilous or not by matching its ADRM against those of known dangerous driving events preserved in a database using a fuzzy attributed map matching technique. The database can automatically augment by including new dangerous driving events that approved any of the predefined danger criteria. A series of experiments with synthetic examples generated by a driving simulator have been conducted to demonstrate the feasibility and rationality of the proposed system.","PeriodicalId":126080,"journal":{"name":"2010 International Conference on Machine Learning and Cybernetics","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126913907","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 : 2010-07-11DOI: 10.1109/ICMLC.2010.5580965
Xiao-xia Zhu, Qun Wang
This paper discusses a hybrid multi-attribute decision making problems (HMADMP) in which attribute weights are unknown and attribute values are given in the forms of real number, interval number, fuzzy number and semantics. Through the process of the attribute values transformation and clarity, proposed the method with the integration information entropy to calculate the comprehensive weights, and established a hybrid multi-attribute decision model based on comprehensive information entropy. Finally, Application examples demonstrate its feasibility and effectiveness.
{"title":"The integration information entropy method for the hybrid multi-attribute decision-making","authors":"Xiao-xia Zhu, Qun Wang","doi":"10.1109/ICMLC.2010.5580965","DOIUrl":"https://doi.org/10.1109/ICMLC.2010.5580965","url":null,"abstract":"This paper discusses a hybrid multi-attribute decision making problems (HMADMP) in which attribute weights are unknown and attribute values are given in the forms of real number, interval number, fuzzy number and semantics. Through the process of the attribute values transformation and clarity, proposed the method with the integration information entropy to calculate the comprehensive weights, and established a hybrid multi-attribute decision model based on comprehensive information entropy. Finally, Application examples demonstrate its feasibility and effectiveness.","PeriodicalId":126080,"journal":{"name":"2010 International Conference on Machine Learning and Cybernetics","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129120481","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 : 2010-07-11DOI: 10.1109/ICMLC.2010.5581012
Dongwen Zhang, Peng Wang, J. Qiu, Yan Jiang
The paper addresses the feature selection based on Neighborhood Rough Set (NRS) used as evaluation function and Ant Colony Optimization (ACO) as generation procedure. A NRS-based measure is employed as heuristic information of ACO. For the weakness of setting a specified value to the size of neighborhood, a new standard deviation based value is advanced to be the size of neighborhood. Four datasets from UCI are used to evaluate the proposed approach and the experimental results show that the approach has a better performance, and could be a practical algorithm to select features from dataset.
{"title":"An improved approach to feature selection","authors":"Dongwen Zhang, Peng Wang, J. Qiu, Yan Jiang","doi":"10.1109/ICMLC.2010.5581012","DOIUrl":"https://doi.org/10.1109/ICMLC.2010.5581012","url":null,"abstract":"The paper addresses the feature selection based on Neighborhood Rough Set (NRS) used as evaluation function and Ant Colony Optimization (ACO) as generation procedure. A NRS-based measure is employed as heuristic information of ACO. For the weakness of setting a specified value to the size of neighborhood, a new standard deviation based value is advanced to be the size of neighborhood. Four datasets from UCI are used to evaluate the proposed approach and the experimental results show that the approach has a better performance, and could be a practical algorithm to select features from dataset.","PeriodicalId":126080,"journal":{"name":"2010 International Conference on Machine Learning and Cybernetics","volume":"129 8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124241377","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 : 2010-07-11DOI: 10.1109/ICMLC.2010.5580978
K. Chang, G. Liang, Meng Sun
A three-level index evaluation system of high-tech enterprises is raised expanding from four main factors-capital, technology, market and management; neural network system is introduced in high-tech enterprises growth-evaluation to strengthen the evaluation; BP network model is established through network training, which is used to carry out growth evaluation. These provide frontier idea for the cross application of mathematics in the field of high-tech enterprise management.
{"title":"Study on strengthen of high-tech enterprises growth-evaluation based on neural network system","authors":"K. Chang, G. Liang, Meng Sun","doi":"10.1109/ICMLC.2010.5580978","DOIUrl":"https://doi.org/10.1109/ICMLC.2010.5580978","url":null,"abstract":"A three-level index evaluation system of high-tech enterprises is raised expanding from four main factors-capital, technology, market and management; neural network system is introduced in high-tech enterprises growth-evaluation to strengthen the evaluation; BP network model is established through network training, which is used to carry out growth evaluation. These provide frontier idea for the cross application of mathematics in the field of high-tech enterprise management.","PeriodicalId":126080,"journal":{"name":"2010 International Conference on Machine Learning and Cybernetics","volume":"104 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123208504","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 : 2010-07-11DOI: 10.1109/ICMLC.2010.5581010
Zhi-Min He, Wing W. Y. Ng, P. Chan, D. Yeung
Steganalysis is a technique to fight against steganography. Different kinds of feature extraction methods have been proposed for blind steganalysis. They have their own advantages when attacking different kinds of steganography. Making a combination of different feature sets will improve the performance of the steganalysis system. However, it will increase the dimensionality of features largely at the same time. Meanwhile, it may have many irrelevant features in the system. A proper feature selection method could decrease the computational complexity and also enhance the performance of the steganalysis. In this paper, we proposed a feature selection method based on the Localized Generalization Error Model (L-GEM) to selection the most relevant feature subset for steganalysis system. The proposed method is compared with two other off-the-shelf feature selection methods. The experimental results show that the proposed method outperforms the other two feature selection methods. The steganalysis with the proposed feature selection method yields a higher average testing accuracy than that of using full set of features.
{"title":"Feature selection for blind steganalysis using localized generalization error model","authors":"Zhi-Min He, Wing W. Y. Ng, P. Chan, D. Yeung","doi":"10.1109/ICMLC.2010.5581010","DOIUrl":"https://doi.org/10.1109/ICMLC.2010.5581010","url":null,"abstract":"Steganalysis is a technique to fight against steganography. Different kinds of feature extraction methods have been proposed for blind steganalysis. They have their own advantages when attacking different kinds of steganography. Making a combination of different feature sets will improve the performance of the steganalysis system. However, it will increase the dimensionality of features largely at the same time. Meanwhile, it may have many irrelevant features in the system. A proper feature selection method could decrease the computational complexity and also enhance the performance of the steganalysis. In this paper, we proposed a feature selection method based on the Localized Generalization Error Model (L-GEM) to selection the most relevant feature subset for steganalysis system. The proposed method is compared with two other off-the-shelf feature selection methods. The experimental results show that the proposed method outperforms the other two feature selection methods. The steganalysis with the proposed feature selection method yields a higher average testing accuracy than that of using full set of features.","PeriodicalId":126080,"journal":{"name":"2010 International Conference on Machine Learning and Cybernetics","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123608484","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 : 2010-07-11DOI: 10.1109/ICMLC.2010.5581043
Hua Li, Gui-Wen Lv, Sumei Zhang, Zhicaho Guo
In this paper, we proposed an extended heuristic algorithm to Fuzzy ID3 using the minimization information entropy and mutual information entropy. Most of the current fuzzy decision trees learning algorithms often select the previously selected attributes for branching. The repeated selection limits the accuracy of training and testing and the structure of decision trees may become complex. Here, we use mutual information to avoid selecting the redundancy attributes in the generation of fuzzy decision tree. The test results show that this method can obtain good performance.
{"title":"Using mutual information for fuzzy decision tree generation","authors":"Hua Li, Gui-Wen Lv, Sumei Zhang, Zhicaho Guo","doi":"10.1109/ICMLC.2010.5581043","DOIUrl":"https://doi.org/10.1109/ICMLC.2010.5581043","url":null,"abstract":"In this paper, we proposed an extended heuristic algorithm to Fuzzy ID3 using the minimization information entropy and mutual information entropy. Most of the current fuzzy decision trees learning algorithms often select the previously selected attributes for branching. The repeated selection limits the accuracy of training and testing and the structure of decision trees may become complex. Here, we use mutual information to avoid selecting the redundancy attributes in the generation of fuzzy decision tree. The test results show that this method can obtain good performance.","PeriodicalId":126080,"journal":{"name":"2010 International Conference on Machine Learning and Cybernetics","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2010-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114079623","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：