Problems where the output varies depending on the input (signal factor levels) are classified as dynamic systems in the Taguchi method. In dynamic systems, if both input and output have only two digital values (0 and 1) with the possibility of committing two types of errors (judging 0 as 1 and 1 as 0), such a problem is called digital system or digital-digital dynamic system. In the digital system, whenever an input signal is 0 or 1, the output is affected by control factors and noise factors, the criterion for judging the output is the threshold value R. If output is smaller than threshold R, output is set as 0 when input signal is 0. Similarly, if output is larger than threshold R, the output is set as 1 when input signal is 1. Hence, two types of error rate are occurred. The purpose of this paper is to apply the Bayesian point estimation method to view the error rates as random variables and optimize the digital system and find the setting value of threshold R for the cases of loss coefficients are unequal. The implementation and the effectiveness of the proposed approach is illustrated through a case study.
{"title":"Robust Design of Digital Dynamic Systems by the Bayesian Point Estimation Method","authors":"Fulchiang Wu","doi":"10.1109/CIIS.2017.33","DOIUrl":"https://doi.org/10.1109/CIIS.2017.33","url":null,"abstract":"Problems where the output varies depending on the input (signal factor levels) are classified as dynamic systems in the Taguchi method. In dynamic systems, if both input and output have only two digital values (0 and 1) with the possibility of committing two types of errors (judging 0 as 1 and 1 as 0), such a problem is called digital system or digital-digital dynamic system. In the digital system, whenever an input signal is 0 or 1, the output is affected by control factors and noise factors, the criterion for judging the output is the threshold value R. If output is smaller than threshold R, output is set as 0 when input signal is 0. Similarly, if output is larger than threshold R, the output is set as 1 when input signal is 1. Hence, two types of error rate are occurred. The purpose of this paper is to apply the Bayesian point estimation method to view the error rates as random variables and optimize the digital system and find the setting value of threshold R for the cases of loss coefficients are unequal. The implementation and the effectiveness of the proposed approach is illustrated through a case study.","PeriodicalId":254342,"journal":{"name":"2017 International Conference on Computing Intelligence and Information System (CIIS)","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124894047","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}
The ACOR algorithm is an Ant Colony Optimization (ACO) extended to continuous domains, and has been used for training neural network. However, when training neural networks, ACOR does not allow for heuristic information like most conventional ACO algorithms do. So in this work we propose a hybrid ACOR algorithm, named h-ACOR, which incorporates the heuristic information into the framework of ACOR for neural network training. The heuristic information in h-ACOR is a gradient vector obtained by computing the partial derivative of error term of the neural network with respect to weight vector. The h-ACOR is tested on training neural networks for pattern classification problems with UCI datasets: zoo, iris and tic-tac-toe. The experiments were carried out using 10-fold cross-validation method, and the results show that: h-ACOR has better performance than ACOR with almost half of convergence generations; and after completely training by h-ACOR, the average classification accuracy of datasets zoo, iris and tic-tac-toe is 92.6% while that of ACOR is 86.6%.
{"title":"A Hybrid ACOR Algorithm for Pattern Classification Neural Network Training","authors":"Zhangming Zhao, Jing Feng, Kunpeng Jing, En Shi","doi":"10.1109/CIIS.2017.35","DOIUrl":"https://doi.org/10.1109/CIIS.2017.35","url":null,"abstract":"The ACOR algorithm is an Ant Colony Optimization (ACO) extended to continuous domains, and has been used for training neural network. However, when training neural networks, ACOR does not allow for heuristic information like most conventional ACO algorithms do. So in this work we propose a hybrid ACOR algorithm, named h-ACOR, which incorporates the heuristic information into the framework of ACOR for neural network training. The heuristic information in h-ACOR is a gradient vector obtained by computing the partial derivative of error term of the neural network with respect to weight vector. The h-ACOR is tested on training neural networks for pattern classification problems with UCI datasets: zoo, iris and tic-tac-toe. The experiments were carried out using 10-fold cross-validation method, and the results show that: h-ACOR has better performance than ACOR with almost half of convergence generations; and after completely training by h-ACOR, the average classification accuracy of datasets zoo, iris and tic-tac-toe is 92.6% while that of ACOR is 86.6%.","PeriodicalId":254342,"journal":{"name":"2017 International Conference on Computing Intelligence and Information System (CIIS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129475463","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}
Salt bridges (Sbs) play an important role in stabilizing protein. Long time molecular dynamics simulation was performed at different temperatures to study salt bridges dynamics and salt bridges networks of Wild Type Lipase (WTL) (Tm=56 °C)and mutant 6B(Tm=78.2 °C). The higher salt bridge persistence value generally means that the Sb is more stable. The persistence values of Lys35-Asp34, Glu171-Arg147 and Asn181-Lys122 in 6B are different from those of WTL with temperature increasing. The improvement on the stability of these Sbs show that they stabilize the protein secondary structure and then strengthen the ability of protein to withstand high temperature. In addition, the mutations A20E and G111D form new Sbs which affect the Sbs networks dynamics of 6B. Due to the mutations, it forms new small Sbs network and enhances the stability of two salt bridge networks. The first Sbs network makes αB, loop and 310-helix closely connect with each other at mutant. The second Sbs network improves the internal interaction of the αB. The third Sbs network enhances the interaction between loops, αD and αE. It is thus clear that the mutations change the stability of Sbs and the Sbs networks which is responsible for increasing thermostability of 6B.
{"title":"The Thermo Stability of Lipase: Salt Bridge and Salt Bridge Network Perspective Based on Long Time Molecular Dynamics Simulation","authors":"Leiyu Zhang, Yanrui Ding","doi":"10.1109/CIIS.2017.34","DOIUrl":"https://doi.org/10.1109/CIIS.2017.34","url":null,"abstract":"Salt bridges (Sbs) play an important role in stabilizing protein. Long time molecular dynamics simulation was performed at different temperatures to study salt bridges dynamics and salt bridges networks of Wild Type Lipase (WTL) (Tm=56 °C)and mutant 6B(Tm=78.2 °C). The higher salt bridge persistence value generally means that the Sb is more stable. The persistence values of Lys35-Asp34, Glu171-Arg147 and Asn181-Lys122 in 6B are different from those of WTL with temperature increasing. The improvement on the stability of these Sbs show that they stabilize the protein secondary structure and then strengthen the ability of protein to withstand high temperature. In addition, the mutations A20E and G111D form new Sbs which affect the Sbs networks dynamics of 6B. Due to the mutations, it forms new small Sbs network and enhances the stability of two salt bridge networks. The first Sbs network makes αB, loop and 310-helix closely connect with each other at mutant. The second Sbs network improves the internal interaction of the αB. The third Sbs network enhances the interaction between loops, αD and αE. It is thus clear that the mutations change the stability of Sbs and the Sbs networks which is responsible for increasing thermostability of 6B.","PeriodicalId":254342,"journal":{"name":"2017 International Conference on Computing Intelligence and Information System (CIIS)","volume":"163 5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127523262","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}
The existing accident alarm systems are based on fixed networks and their structure cannot be changed. They cannot prevent secondary accidents timely or get the help from adjacent drivers or pedestrians. A dynamic alarm system, named Traffic Emergency Detection and Alarm System, is proposed in this paper. The system not only collects the accident information, but also automatically forms a dynamic network, sends out alarm messages to adjacent drivers or pedestrians, and combines different resources or objects to provide an efficient medical aid. In addition, an alarm targets selection algorithm is proposed, which is based on a dynamic Analytic Hierarchy Process (AHP) method. Two cases are analyzed in detail, which cover the normal and special circumstances. They illustrate that the dynamic AHP method can successfully adapt to selecting alarm targets in different scenarios.
{"title":"Alarm Targets Selection Algorithm Based on Dynamic AHP Method","authors":"X. Chen, Zhiting Lin","doi":"10.1109/CIIS.2017.62","DOIUrl":"https://doi.org/10.1109/CIIS.2017.62","url":null,"abstract":"The existing accident alarm systems are based on fixed networks and their structure cannot be changed. They cannot prevent secondary accidents timely or get the help from adjacent drivers or pedestrians. A dynamic alarm system, named Traffic Emergency Detection and Alarm System, is proposed in this paper. The system not only collects the accident information, but also automatically forms a dynamic network, sends out alarm messages to adjacent drivers or pedestrians, and combines different resources or objects to provide an efficient medical aid. In addition, an alarm targets selection algorithm is proposed, which is based on a dynamic Analytic Hierarchy Process (AHP) method. Two cases are analyzed in detail, which cover the normal and special circumstances. They illustrate that the dynamic AHP method can successfully adapt to selecting alarm targets in different scenarios.","PeriodicalId":254342,"journal":{"name":"2017 International Conference on Computing Intelligence and Information System (CIIS)","volume":"66 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128929139","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}
An information system as the application background, based on the idea of PVM virtual machine, a supporting multilingual customizable model of workflow system is proposed. The key point is the analysis of a kind of language describing the flow processes and its resolution process. Such integration can be customized workflow engine and flexible information system modeling languages can be used to make the implementation process of clarity. The difficulty of the development and maintenance for information system is effectively reduced, with the scalability and flexibility enhanced.
{"title":"Customizable Modeling Method of Workflow Engine Used in Information System","authors":"Qing Yu, Xi-Wu Gu","doi":"10.1109/CIIS.2017.43","DOIUrl":"https://doi.org/10.1109/CIIS.2017.43","url":null,"abstract":"An information system as the application background, based on the idea of PVM virtual machine, a supporting multilingual customizable model of workflow system is proposed. The key point is the analysis of a kind of language describing the flow processes and its resolution process. Such integration can be customized workflow engine and flexible information system modeling languages can be used to make the implementation process of clarity. The difficulty of the development and maintenance for information system is effectively reduced, with the scalability and flexibility enhanced.","PeriodicalId":254342,"journal":{"name":"2017 International Conference on Computing Intelligence and Information System (CIIS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130760952","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}
Fine-grained product feature extraction is the most important task in opinion mining. To realize the fine-grained product feature extraction in Chinese reviews, three main tasks have been solved in this paper. Firstly, we propose a dependency parsing based method to directly extract the explicit feature-opinion pairs. Then, by analyzing the characteristics of two synonyms features and the relations with opinion words, we calculate the similarities to cluster features. Finally, we propose a novel implicit feature extraction method by combining review context information and two kind opinions to extract implicit features. Experiments show that the dependency parsing based method can get high precision, by considering verbs as product feature can improve the recall obviously. Besides, several proven pruning strategies can improve the accuracy. The comparison demonstrates that our implicit feature extraction method outperforms existing method, and feature clustering before implicit feature mining can get better results.
{"title":"Fine-Grained Product Feature Extraction in Chinese Reviews","authors":"Hanqian Wu, Tao Liu, Jue Xie","doi":"10.1109/CIIS.2017.53","DOIUrl":"https://doi.org/10.1109/CIIS.2017.53","url":null,"abstract":"Fine-grained product feature extraction is the most important task in opinion mining. To realize the fine-grained product feature extraction in Chinese reviews, three main tasks have been solved in this paper. Firstly, we propose a dependency parsing based method to directly extract the explicit feature-opinion pairs. Then, by analyzing the characteristics of two synonyms features and the relations with opinion words, we calculate the similarities to cluster features. Finally, we propose a novel implicit feature extraction method by combining review context information and two kind opinions to extract implicit features. Experiments show that the dependency parsing based method can get high precision, by considering verbs as product feature can improve the recall obviously. Besides, several proven pruning strategies can improve the accuracy. The comparison demonstrates that our implicit feature extraction method outperforms existing method, and feature clustering before implicit feature mining can get better results.","PeriodicalId":254342,"journal":{"name":"2017 International Conference on Computing Intelligence and Information System (CIIS)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132027972","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}
Based on the analysis of the typical cabin structure, the dismantling of the target unit and the division of the damage level, the damage assessment criteria, the index system and the quantitative calculation model of the naval vessel bomb were established. And provide the basis for the evaluation of the combat effectiveness of the surface warships, so as to optimize the combat command and decision-making and rational use of shooting methods to provide support.
{"title":"Evaluation Method of the Damage Effect Produced by Naval Gun Armor-Piercing Blast Shell to Vessel Cabin","authors":"Dongyan Sun","doi":"10.1109/ciis.2017.29","DOIUrl":"https://doi.org/10.1109/ciis.2017.29","url":null,"abstract":"Based on the analysis of the typical cabin structure, the dismantling of the target unit and the division of the damage level, the damage assessment criteria, the index system and the quantitative calculation model of the naval vessel bomb were established. And provide the basis for the evaluation of the combat effectiveness of the surface warships, so as to optimize the combat command and decision-making and rational use of shooting methods to provide support.","PeriodicalId":254342,"journal":{"name":"2017 International Conference on Computing Intelligence and Information System (CIIS)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127937424","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}
W. Ying, Shiyun Chen, Bingshen Wu, Yuehong Xie, Yu Wu
The multi-objective evolutionary algorithm based on decomposition (MOEA/D) has shown remarkable performance for multi-objective optimization problems (MOPs). However, MOEA/D still consumes long time to solve MOPs with computationally intensive objective functions. This paper proposes two distributed parallel MOEA/Ds based on the popular distributed framework, Spark, to further reduce the running time of the sequential MOEA/D for MOPs. The first entirely evolved MOEA/D evolves an entire population, while the second partially evolved MOEA/D based on Spark evolves a partial subpopulation equal in size to a partition in each transformation-action process. Experimental results on DTLZ benchmark MOPs with three objectives indicate that both distributed MOEA/Ds on Spark obtains better speedup than the distributed MOEA/Ds on MapReduce and achieve the quality of solutions similar to the sequential MOEA/D.
{"title":"Distributed Parellel MOEA/D on Spark","authors":"W. Ying, Shiyun Chen, Bingshen Wu, Yuehong Xie, Yu Wu","doi":"10.1109/CIIS.2017.12","DOIUrl":"https://doi.org/10.1109/CIIS.2017.12","url":null,"abstract":"The multi-objective evolutionary algorithm based on decomposition (MOEA/D) has shown remarkable performance for multi-objective optimization problems (MOPs). However, MOEA/D still consumes long time to solve MOPs with computationally intensive objective functions. This paper proposes two distributed parallel MOEA/Ds based on the popular distributed framework, Spark, to further reduce the running time of the sequential MOEA/D for MOPs. The first entirely evolved MOEA/D evolves an entire population, while the second partially evolved MOEA/D based on Spark evolves a partial subpopulation equal in size to a partition in each transformation-action process. Experimental results on DTLZ benchmark MOPs with three objectives indicate that both distributed MOEA/Ds on Spark obtains better speedup than the distributed MOEA/Ds on MapReduce and achieve the quality of solutions similar to the sequential MOEA/D.","PeriodicalId":254342,"journal":{"name":"2017 International Conference on Computing Intelligence and Information System (CIIS)","volume":"131 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125492004","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}
Point clouds which generated from spinning multi-laser sensors are sparse and with uneven density. When dealing with such point clouds, the traditional plane extraction algorithm encounters contradicting issues: speed and accuracy. This paper presents a plane extraction method based on the Randomized Hough Transform. A spherical accumulator model is used to decrease computational costs and a point selection method is presented to resolve the difficulty caused by uneven density. In addition, a standard deviation threshold of the inner points is set to exclude the wrong detections. The algorithm has a good application for plane extraction in 3D sparse point cloud. Experiments shown that using our method we were able to detect plane with a better accuracy than traditional methods.
{"title":"A Plane Extraction Method Based on the Randomized Hough Transform","authors":"Xiaoqing Wang, Chenjing Ding, Yongping Wang, Xingqun Zhao","doi":"10.1109/CIIS.2017.32","DOIUrl":"https://doi.org/10.1109/CIIS.2017.32","url":null,"abstract":"Point clouds which generated from spinning multi-laser sensors are sparse and with uneven density. When dealing with such point clouds, the traditional plane extraction algorithm encounters contradicting issues: speed and accuracy. This paper presents a plane extraction method based on the Randomized Hough Transform. A spherical accumulator model is used to decrease computational costs and a point selection method is presented to resolve the difficulty caused by uneven density. In addition, a standard deviation threshold of the inner points is set to exclude the wrong detections. The algorithm has a good application for plane extraction in 3D sparse point cloud. Experiments shown that using our method we were able to detect plane with a better accuracy than traditional methods.","PeriodicalId":254342,"journal":{"name":"2017 International Conference on Computing Intelligence and Information System (CIIS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122662461","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}
The deep neural networks (DNN) based speech enhancements is a hot topic in machine learning and speech enhancement application. Even with deep neural network, it is still hard to improve the speech quality on noisy and reverberant conditions. For machine learning based system, auditory feature extraction becomes the key point in speech enhancement and recognition. In this paper, we proposed a speech enhancement framework based on an auditory-based monaural feature, which model the function of human hearing auditory system. The auditory based feature is extracted from the data passing the gammatone filter banks, which has more detail on low frequency than normal filters. Systemic tests show the better performance of the proposed auditory based monaural feature than the mel-frequency cepstral coefficients (MFCC) in noise and reverberant environment.
{"title":"An Auditory-Based Monaural Feature for Noisy and Reverberant Speech Enhancement","authors":"Yi-jiao Jiang, Runsheng Liu, Ya Bai","doi":"10.1109/CIIS.2017.23","DOIUrl":"https://doi.org/10.1109/CIIS.2017.23","url":null,"abstract":"The deep neural networks (DNN) based speech enhancements is a hot topic in machine learning and speech enhancement application. Even with deep neural network, it is still hard to improve the speech quality on noisy and reverberant conditions. For machine learning based system, auditory feature extraction becomes the key point in speech enhancement and recognition. In this paper, we proposed a speech enhancement framework based on an auditory-based monaural feature, which model the function of human hearing auditory system. The auditory based feature is extracted from the data passing the gammatone filter banks, which has more detail on low frequency than normal filters. Systemic tests show the better performance of the proposed auditory based monaural feature than the mel-frequency cepstral coefficients (MFCC) in noise and reverberant environment.","PeriodicalId":254342,"journal":{"name":"2017 International Conference on Computing Intelligence and Information System (CIIS)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123112751","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：