Pub Date : 2017-11-01DOI: 10.1109/ISKE.2017.8258845
Shu Li, Yun Li
Machine learning has been widely used in security related applications, such as spam filter, network intrusion detection. In machine learning process, the test set and the training set usually have the same probability distribution and through the information of learning the training set, the malicious samples in the machine learning algorithm can usually be correctly classified. However, the classification algorithm has neglected the classification under adversarial environment, so instead they will modify the features of test data in order to spoof the classifier so as to escape its detection. In this paper, we will consider to modify the feature value of the test samples in accordance with attack algorithm proposed by Battista Biggio and further improve the algorithm. As each feature has a range of independent constraints, so the algorithm should be transformed into a constrained optimization problem. This is done in order to make the original sample modify the smaller distance so as to escape the detection of the classifier, while also improve the convergence rate during the generation of adversarial samples.
{"title":"Complex-based optimization strategy for evasion attack","authors":"Shu Li, Yun Li","doi":"10.1109/ISKE.2017.8258845","DOIUrl":"https://doi.org/10.1109/ISKE.2017.8258845","url":null,"abstract":"Machine learning has been widely used in security related applications, such as spam filter, network intrusion detection. In machine learning process, the test set and the training set usually have the same probability distribution and through the information of learning the training set, the malicious samples in the machine learning algorithm can usually be correctly classified. However, the classification algorithm has neglected the classification under adversarial environment, so instead they will modify the features of test data in order to spoof the classifier so as to escape its detection. In this paper, we will consider to modify the feature value of the test samples in accordance with attack algorithm proposed by Battista Biggio and further improve the algorithm. As each feature has a range of independent constraints, so the algorithm should be transformed into a constrained optimization problem. This is done in order to make the original sample modify the smaller distance so as to escape the detection of the classifier, while also improve the convergence rate during the generation of adversarial samples.","PeriodicalId":208009,"journal":{"name":"2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114758662","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 : 2017-11-01DOI: 10.1109/ISKE.2017.8258717
Xu Chen, Long Hong, Guofang Huang
With the continuous development of the Internet and information technology, data has penetrated into every area of today's industry and business functions. Now, data has been already one of the most valuable assets in the Internet and a core element of a company's competitiveness. There seems to have endless data on the Internet, then most of it cannot create value. When we face data mining, wo need to complete a lot of data cleaning tasks. Nowadays, the rapid development of machine learning, especially the deep of learning, has made excellent achievements in natural language processing and image recognition. The paper combines multiple strong learning machine to complete the data learning tasks and image recognition based on ensemble learning, thereby reduce the pressure on the server storage and investment of resource.
{"title":"Ensemble learning for image recognition","authors":"Xu Chen, Long Hong, Guofang Huang","doi":"10.1109/ISKE.2017.8258717","DOIUrl":"https://doi.org/10.1109/ISKE.2017.8258717","url":null,"abstract":"With the continuous development of the Internet and information technology, data has penetrated into every area of today's industry and business functions. Now, data has been already one of the most valuable assets in the Internet and a core element of a company's competitiveness. There seems to have endless data on the Internet, then most of it cannot create value. When we face data mining, wo need to complete a lot of data cleaning tasks. Nowadays, the rapid development of machine learning, especially the deep of learning, has made excellent achievements in natural language processing and image recognition. The paper combines multiple strong learning machine to complete the data learning tasks and image recognition based on ensemble learning, thereby reduce the pressure on the server storage and investment of resource.","PeriodicalId":208009,"journal":{"name":"2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121956071","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 : 2017-11-01DOI: 10.1109/ISKE.2017.8258825
Minyang Kang, Yang Liu, Yijie Ren, Yijing Zhao, Zheng Zheng
UAV(Unmanned Aerial Vehicle) needs to accomplish its task with obstacle avoidance. However, uncertainties in the actual complex flight environment affect the application of UAV. In consideration of the error of UAV's position estimation, this paper attempts to evaluate the robustness which is measured by the safety degree of the path. UAV path planning algorithms, including A-Star, BLP(bi-level programming based algorithm), PSO(Particle Swarm Optimization) and RRT(Rapid-exploring Random Trees), are selected for the empirical study. Results demonstrate that RRT and BLP behave much better than A∗ and PSO, considering variance and scenario complexity. RRT algorithm performs better in the simpler scenario and larger variance and BLP algorithm is more robust in the case of low variance.
{"title":"An empirical study on robustness of UAV path planning algorithms considering position uncertainty","authors":"Minyang Kang, Yang Liu, Yijie Ren, Yijing Zhao, Zheng Zheng","doi":"10.1109/ISKE.2017.8258825","DOIUrl":"https://doi.org/10.1109/ISKE.2017.8258825","url":null,"abstract":"UAV(Unmanned Aerial Vehicle) needs to accomplish its task with obstacle avoidance. However, uncertainties in the actual complex flight environment affect the application of UAV. In consideration of the error of UAV's position estimation, this paper attempts to evaluate the robustness which is measured by the safety degree of the path. UAV path planning algorithms, including A-Star, BLP(bi-level programming based algorithm), PSO(Particle Swarm Optimization) and RRT(Rapid-exploring Random Trees), are selected for the empirical study. Results demonstrate that RRT and BLP behave much better than A∗ and PSO, considering variance and scenario complexity. RRT algorithm performs better in the simpler scenario and larger variance and BLP algorithm is more robust in the case of low variance.","PeriodicalId":208009,"journal":{"name":"2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129254586","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 : 2017-11-01DOI: 10.1109/ISKE.2017.8258739
Zhen Zhang, Xinyue Kou, W. Yu
Two-sided matching problems exist widely in human beings' daily life. In this paper, two-sided matching decision making problems with heterogeneous incomplete preference relations are investigated. In order to obtain the optimal matching between matching objects on both sides, the priority weight vectors are firstly derived from each matching object's incomplete fuzzy or multiplicative preference relation over matching objects on the other side. Based on the priority weight vector, each matching object's satisfaction degrees over matching objects on the other side are calculated, based on which a bi-objective linear binary programming model is constructed and solved to determine the optimal matching. Finally, an example for employee-position matching is provided to illustrate the proposed approach.
{"title":"Two-sided matching decision making based on heterogeneous incomplete preference relations","authors":"Zhen Zhang, Xinyue Kou, W. Yu","doi":"10.1109/ISKE.2017.8258739","DOIUrl":"https://doi.org/10.1109/ISKE.2017.8258739","url":null,"abstract":"Two-sided matching problems exist widely in human beings' daily life. In this paper, two-sided matching decision making problems with heterogeneous incomplete preference relations are investigated. In order to obtain the optimal matching between matching objects on both sides, the priority weight vectors are firstly derived from each matching object's incomplete fuzzy or multiplicative preference relation over matching objects on the other side. Based on the priority weight vector, each matching object's satisfaction degrees over matching objects on the other side are calculated, based on which a bi-objective linear binary programming model is constructed and solved to determine the optimal matching. Finally, an example for employee-position matching is provided to illustrate the proposed approach.","PeriodicalId":208009,"journal":{"name":"2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129272522","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 : 2017-11-01DOI: 10.1109/ISKE.2017.8258821
Yupeng Ding, Hongjun Li, Zhengyu Li
In order to solve the confusion of input data, an algorithm of human action recognition based on packet convolution neural network is proposed. The two-layer wavelet combined with the mean square error method is used to group the samples, and then study the features in the case of guaranteeing the grouping error. The algorithm is tested on the video library and compared with the traditional convolution neural network algorithm. The experimental results show that the proposed algorithm has a significant improvement in the subjective and objective performance compared with the similar algorithm, and the success rate has been greatly improved.
{"title":"Human motion recognition based on packet convolution neural network","authors":"Yupeng Ding, Hongjun Li, Zhengyu Li","doi":"10.1109/ISKE.2017.8258821","DOIUrl":"https://doi.org/10.1109/ISKE.2017.8258821","url":null,"abstract":"In order to solve the confusion of input data, an algorithm of human action recognition based on packet convolution neural network is proposed. The two-layer wavelet combined with the mean square error method is used to group the samples, and then study the features in the case of guaranteeing the grouping error. The algorithm is tested on the video library and compared with the traditional convolution neural network algorithm. The experimental results show that the proposed algorithm has a significant improvement in the subjective and objective performance compared with the similar algorithm, and the success rate has been greatly improved.","PeriodicalId":208009,"journal":{"name":"2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127656210","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 growing of crowdsourcing systems, class labels for supervised learning can be easily obtained from crowdsourcing platforms. To deal with the problem that labels obtained from crowds are usually noisy due to imperfect reliability of non-expert workers, we let multiple workers provide labels for the same object. Then, true labels of the labeled object are estimated through ground truth inference algorithms. The inferred integrated labels are expected to be of high quality. In this paper, we propose a novel ground truth inference algorithm based on EM algorithm, which not only infers the true labels of the instances but also simultaneously estimates the reliability of each worker and the difficulty of each instance. Experimental results on seven real-world crowdsourcing datasets show that our proposed algorithm outperforms eight state-of-the art algorithms.
{"title":"A robust inference algorithm for crowd sourced categorization","authors":"Ming Wu, Qianmu Li, Jing Zhang, Shicheng Cui, Deqiang Li, Yong Qi","doi":"10.1109/ISKE.2017.8258809","DOIUrl":"https://doi.org/10.1109/ISKE.2017.8258809","url":null,"abstract":"With the rapid growing of crowdsourcing systems, class labels for supervised learning can be easily obtained from crowdsourcing platforms. To deal with the problem that labels obtained from crowds are usually noisy due to imperfect reliability of non-expert workers, we let multiple workers provide labels for the same object. Then, true labels of the labeled object are estimated through ground truth inference algorithms. The inferred integrated labels are expected to be of high quality. In this paper, we propose a novel ground truth inference algorithm based on EM algorithm, which not only infers the true labels of the instances but also simultaneously estimates the reliability of each worker and the difficulty of each instance. Experimental results on seven real-world crowdsourcing datasets show that our proposed algorithm outperforms eight state-of-the art algorithms.","PeriodicalId":208009,"journal":{"name":"2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124408589","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 : 2017-11-01DOI: 10.1109/ISKE.2017.8258726
Yuanyuan Zhao, F. Qin
In this work, we partially study a problem which generalizes and comes from an open problem related to law of importation and suggested by international conference on fuzzy set theory and applications 8th in 2006. In fact, this problem has also partially been investigated in [16]. The obtained results in our paper dramatically extend the aforementioned ones. In detail, we characterizes all fuzzy implications with a continuous α-natural negation which satisfy the law of importation w.r.t a fixed uninorm U, where U is a uninorm continuous in (0,1)2.
{"title":"Characterization of fuzzy implication functions with a continuous α-natural negation satisfying the law of importation with a given uninorm-revisited","authors":"Yuanyuan Zhao, F. Qin","doi":"10.1109/ISKE.2017.8258726","DOIUrl":"https://doi.org/10.1109/ISKE.2017.8258726","url":null,"abstract":"In this work, we partially study a problem which generalizes and comes from an open problem related to law of importation and suggested by international conference on fuzzy set theory and applications 8th in 2006. In fact, this problem has also partially been investigated in [16]. The obtained results in our paper dramatically extend the aforementioned ones. In detail, we characterizes all fuzzy implications with a continuous α-natural negation which satisfy the law of importation w.r.t a fixed uninorm U, where U is a uninorm continuous in (0,1)2.","PeriodicalId":208009,"journal":{"name":"2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)","volume":"216 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115832971","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 : 2017-11-01DOI: 10.1109/ISKE.2017.8258763
S. Wan, Zhen Jin, Feng Wang
Pythagorean fuzzy set (PFS), as an extension of intuitionistic fuzzy set, has received great attention in decision field. How to rank Pythagorean fuzzy numbers (PFNs) is a critical issue during the decision process. Thus, this paper focuses on the ranking method for PFNs. The main works are outlined as follows: (1) Existing ranking methods for PFNs are reviewed. Some examples are proposed to illustrate their limitations. (2) To overcome these limitations, the concepts of knowledge measure and information reliability of PFN are presented to describe the amount and quality of information of PFNs. It is comprehensive to involve the information of positive ideal point, negative ideal point and fuzzy point. (3) Motivated by the concept of relative closeness degree, an arc-length based relative closeness degree of PFN is proposed and interpreted geometrically. Moreover, the arc-length based relative closeness degree is simple and convenient for calculation. (4) A ranking method for PFNs is put forward on the basis of knowledge measure, information reliability and an arc-length based relative closeness degree.
{"title":"A new ranking method for Pythagorean fuzzy numbers","authors":"S. Wan, Zhen Jin, Feng Wang","doi":"10.1109/ISKE.2017.8258763","DOIUrl":"https://doi.org/10.1109/ISKE.2017.8258763","url":null,"abstract":"Pythagorean fuzzy set (PFS), as an extension of intuitionistic fuzzy set, has received great attention in decision field. How to rank Pythagorean fuzzy numbers (PFNs) is a critical issue during the decision process. Thus, this paper focuses on the ranking method for PFNs. The main works are outlined as follows: (1) Existing ranking methods for PFNs are reviewed. Some examples are proposed to illustrate their limitations. (2) To overcome these limitations, the concepts of knowledge measure and information reliability of PFN are presented to describe the amount and quality of information of PFNs. It is comprehensive to involve the information of positive ideal point, negative ideal point and fuzzy point. (3) Motivated by the concept of relative closeness degree, an arc-length based relative closeness degree of PFN is proposed and interpreted geometrically. Moreover, the arc-length based relative closeness degree is simple and convenient for calculation. (4) A ranking method for PFNs is put forward on the basis of knowledge measure, information reliability and an arc-length based relative closeness degree.","PeriodicalId":208009,"journal":{"name":"2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121959932","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 : 2017-11-01DOI: 10.1109/ISKE.2017.8258790
Kamal Bashir, Tianrui Li, Chubato Wondaferaw Yohannese, M. Yahaya
Software Defect Prediction (SDP) proposes to define the exposure of software to defect by building prediction models through using defect data and the software metrics with several learning algorithms which aid in identifying potentially faulty program modules, thus leading to optimal resource allocation and utilization. However, the quality of data and robustness of classifiers affect the accuracy of prediction for these models of classification compromised by data quality such as high dimensionality, class imbalance and the presence of noise in the software defect datasets. This paper presents a combined framework to enhance SDP models in which we use ranker Feature Selection (FS) techniques, Data Sampling (DS) and Iterative-Partition Filter (IPF) to defeat high dimensionality, class imbalance and noisy, respectively. The experimental results confirm that the proposed framework is effective for SDP.
{"title":"Enhancing software defect prediction using supervised-learning based framework","authors":"Kamal Bashir, Tianrui Li, Chubato Wondaferaw Yohannese, M. Yahaya","doi":"10.1109/ISKE.2017.8258790","DOIUrl":"https://doi.org/10.1109/ISKE.2017.8258790","url":null,"abstract":"Software Defect Prediction (SDP) proposes to define the exposure of software to defect by building prediction models through using defect data and the software metrics with several learning algorithms which aid in identifying potentially faulty program modules, thus leading to optimal resource allocation and utilization. However, the quality of data and robustness of classifiers affect the accuracy of prediction for these models of classification compromised by data quality such as high dimensionality, class imbalance and the presence of noise in the software defect datasets. This paper presents a combined framework to enhance SDP models in which we use ranker Feature Selection (FS) techniques, Data Sampling (DS) and Iterative-Partition Filter (IPF) to defeat high dimensionality, class imbalance and noisy, respectively. The experimental results confirm that the proposed framework is effective for SDP.","PeriodicalId":208009,"journal":{"name":"2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115124419","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 : 2017-11-01DOI: 10.1109/ISKE.2017.8258742
Yingmin Guo, Kuan-kuan Zhao, Wei Wang
We proposed the relation between implicative pseudo-filter and Boolean filter of pseudo BCK algebras with condition (pP) or bounded pseudo BCK algebras with condition (pP) and partly solved open problems that "In pseudo BCK algebra or bounded pseudo BCK algebra, is the notion of implicative pseudo-filter equivalent to the notion of Boolean filter?" and "A pseudo BCK algebra is an implicative pseudo BCK algebras if and only if every pseudo-filters of it is Boolean filter (or implicative pseudo-filters).
{"title":"Open problems on implicative pseudo-filters and boolean filters","authors":"Yingmin Guo, Kuan-kuan Zhao, Wei Wang","doi":"10.1109/ISKE.2017.8258742","DOIUrl":"https://doi.org/10.1109/ISKE.2017.8258742","url":null,"abstract":"We proposed the relation between implicative pseudo-filter and Boolean filter of pseudo BCK algebras with condition (pP) or bounded pseudo BCK algebras with condition (pP) and partly solved open problems that \"In pseudo BCK algebra or bounded pseudo BCK algebra, is the notion of implicative pseudo-filter equivalent to the notion of Boolean filter?\" and \"A pseudo BCK algebra is an implicative pseudo BCK algebras if and only if every pseudo-filters of it is Boolean filter (or implicative pseudo-filters).","PeriodicalId":208009,"journal":{"name":"2017 12th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)","volume":"125 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122720976","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：