Pub Date : 2014-12-01DOI: 10.1109/HIS.2014.7086204
I. Fliss, M. Tagina
The presence of faults in complex systems can cause serious damage and even generate fatal situations. Proper and timely diagnosis of behavior of such systems is then crucial to detect the presence of eventual faults and isolate their causes. In this context, this paper describes a new intelligent approach to diagnose multiple faults in complex systems. This approach is based on the combination of a fuzzy system optimized by cultural algorithm and causal reasoning. The ongoing experiments focus on a simulation of the three-tank hydraulic system, a benchmark in the diagnosis domain.
{"title":"Hybrid intelligent approach to diagnose multiple faults in complex systems","authors":"I. Fliss, M. Tagina","doi":"10.1109/HIS.2014.7086204","DOIUrl":"https://doi.org/10.1109/HIS.2014.7086204","url":null,"abstract":"The presence of faults in complex systems can cause serious damage and even generate fatal situations. Proper and timely diagnosis of behavior of such systems is then crucial to detect the presence of eventual faults and isolate their causes. In this context, this paper describes a new intelligent approach to diagnose multiple faults in complex systems. This approach is based on the combination of a fuzzy system optimized by cultural algorithm and causal reasoning. The ongoing experiments focus on a simulation of the three-tank hydraulic system, a benchmark in the diagnosis domain.","PeriodicalId":161103,"journal":{"name":"2014 14th International Conference on Hybrid Intelligent Systems","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126441797","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 : 2014-12-01DOI: 10.1109/HIS.2014.7086205
Rim Afdhal, R. Ejbali, M. Zaied, C. Amar
This paper focuses on the issue of emotion recognition. It describes an emotion recognition system based on facial expression which contains four steps: detection of face's elements, localization of feature points, their tracking during a movie and facial expression classification. The first step is realized by the famous Viola and Jones algorithm. To localize feature points we have developed an automatic and easy method. To track them we used the optical flow. Finally the classification step is based on wavelet network using Fast Wavelet Transform FWT. The experimental results demonstrated the efficiency of our system.
{"title":"Emotion recognition using features distances classified by wavelets network and trained by fast wavelets transform","authors":"Rim Afdhal, R. Ejbali, M. Zaied, C. Amar","doi":"10.1109/HIS.2014.7086205","DOIUrl":"https://doi.org/10.1109/HIS.2014.7086205","url":null,"abstract":"This paper focuses on the issue of emotion recognition. It describes an emotion recognition system based on facial expression which contains four steps: detection of face's elements, localization of feature points, their tracking during a movie and facial expression classification. The first step is realized by the famous Viola and Jones algorithm. To localize feature points we have developed an automatic and easy method. To track them we used the optical flow. Finally the classification step is based on wavelet network using Fast Wavelet Transform FWT. The experimental results demonstrated the efficiency of our system.","PeriodicalId":161103,"journal":{"name":"2014 14th International Conference on Hybrid Intelligent Systems","volume":"166 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134315884","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 : 2014-12-01DOI: 10.1109/HIS.2014.7086198
Olayemi Olawumi, Keijo Haataja, Mikko Asikainen, Niko Vidgren, Pekka J. Toivanen
In this paper, three practical attacks against ZigBee security are carried out in our laboratory environment. The attack scenarios are based on utilizing several vulnerabilities found from the main security components of ZigBee technology. The first attack is based on discovering all ZigBee-enabled networks within range as well as the configurations of the corresponding ZigBee-enabled devices: This vital and fundamental basic information can be used for performing further and more severe attacks against the discovered ZigBee-enabled devices/networks. The second attack can be seen as an extension to the first attack and thus the prerequisite for it is the successful completion of the first attack. In the second attack, an attacker eavesdrops on the unencrypted or encrypted traffic of a ZigBee-enabled network in order to obtain and utilize any sensitive/useful information. The third attack is based on replaying (re-transmitting) the captured data as if the original sender is sending the data again. To keep this attack extremely simple, straightforward, and practical, we decided to devise and implement it without having a Man-In-The-Middle (MITM) between the victim devices, since the presence of the MITM would have made the attack very difficult to implement in practice, thus giving it only a theoretical relevance. Indeed, we demonstrate with experimental figures that attacks against ZigBee-enabled devices become practical by using our three attack scenarios. In addition, countermeasures that render the attacks impractical, although not totally eliminating their potential danger, are devised. Moreover, some new ideas that will be used in our future research work are proposed.
{"title":"Three practical attacks against ZigBee security: Attack scenario definitions, practical experiments, countermeasures, and lessons learned","authors":"Olayemi Olawumi, Keijo Haataja, Mikko Asikainen, Niko Vidgren, Pekka J. Toivanen","doi":"10.1109/HIS.2014.7086198","DOIUrl":"https://doi.org/10.1109/HIS.2014.7086198","url":null,"abstract":"In this paper, three practical attacks against ZigBee security are carried out in our laboratory environment. The attack scenarios are based on utilizing several vulnerabilities found from the main security components of ZigBee technology. The first attack is based on discovering all ZigBee-enabled networks within range as well as the configurations of the corresponding ZigBee-enabled devices: This vital and fundamental basic information can be used for performing further and more severe attacks against the discovered ZigBee-enabled devices/networks. The second attack can be seen as an extension to the first attack and thus the prerequisite for it is the successful completion of the first attack. In the second attack, an attacker eavesdrops on the unencrypted or encrypted traffic of a ZigBee-enabled network in order to obtain and utilize any sensitive/useful information. The third attack is based on replaying (re-transmitting) the captured data as if the original sender is sending the data again. To keep this attack extremely simple, straightforward, and practical, we decided to devise and implement it without having a Man-In-The-Middle (MITM) between the victim devices, since the presence of the MITM would have made the attack very difficult to implement in practice, thus giving it only a theoretical relevance. Indeed, we demonstrate with experimental figures that attacks against ZigBee-enabled devices become practical by using our three attack scenarios. In addition, countermeasures that render the attacks impractical, although not totally eliminating their potential danger, are devised. Moreover, some new ideas that will be used in our future research work are proposed.","PeriodicalId":161103,"journal":{"name":"2014 14th International Conference on Hybrid Intelligent Systems","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134538192","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 : 2014-12-01DOI: 10.1109/HIS.2014.7086178
D. Kriksciuniene, Marius Liutvinavicius, V. Sakalauskas, Darius Tamasauskas
The amount of data in financial institutions is growing rapidly and the subject of “big data” has become an urgent trend. The “big data” phenomenon brings challenge to empower analytical methods for enhanced scope. At the same time the big data composed from various sources opens new possibilities to capitalize data research. The article investigates the anomalies in big data used by financial institutions. It proposes the model designed for exploring dynamics and detecting anomalous behavior of bank customers. The experimental screening on bank customers' big data shows significant time and necessary calculation steps reduction for detecting suspicious operations.
{"title":"Research of customer behavior anomalies in big financial data","authors":"D. Kriksciuniene, Marius Liutvinavicius, V. Sakalauskas, Darius Tamasauskas","doi":"10.1109/HIS.2014.7086178","DOIUrl":"https://doi.org/10.1109/HIS.2014.7086178","url":null,"abstract":"The amount of data in financial institutions is growing rapidly and the subject of “big data” has become an urgent trend. The “big data” phenomenon brings challenge to empower analytical methods for enhanced scope. At the same time the big data composed from various sources opens new possibilities to capitalize data research. The article investigates the anomalies in big data used by financial institutions. It proposes the model designed for exploring dynamics and detecting anomalous behavior of bank customers. The experimental screening on bank customers' big data shows significant time and necessary calculation steps reduction for detecting suspicious operations.","PeriodicalId":161103,"journal":{"name":"2014 14th International Conference on Hybrid Intelligent Systems","volume":"243 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114579087","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 : 2014-12-01DOI: 10.1109/HIS.2014.7086212
A. Madureira, Bruno Cunha, J. Pereira, S. Gomes, I. Pereira, Jorge M. Santos, A. Abraham
User modeling and user adaptive interaction has become a central research issue to understand users as they interact with technology. The importance of the development of well adapted interfaces to several kinds of users and the differences that characterize them is the basis of the successful interaction. User Personas is a technique that allows the discovery and definition of the archetype users of a system. With that knowledge, the system should shape itself, inferring the user expertise to provide its users with the best possible experience. In this paper, an architecture that combines User Personas and a dynamic, evolving system is proposed, along with an evaluation by its target users. The proposed system is able to infer the user and its matching Persona, and keeps shaping itself in parallel with the user's discovery of the system.
{"title":"Using personas for supporting user modeling on scheduling systems","authors":"A. Madureira, Bruno Cunha, J. Pereira, S. Gomes, I. Pereira, Jorge M. Santos, A. Abraham","doi":"10.1109/HIS.2014.7086212","DOIUrl":"https://doi.org/10.1109/HIS.2014.7086212","url":null,"abstract":"User modeling and user adaptive interaction has become a central research issue to understand users as they interact with technology. The importance of the development of well adapted interfaces to several kinds of users and the differences that characterize them is the basis of the successful interaction. User Personas is a technique that allows the discovery and definition of the archetype users of a system. With that knowledge, the system should shape itself, inferring the user expertise to provide its users with the best possible experience. In this paper, an architecture that combines User Personas and a dynamic, evolving system is proposed, along with an evaluation by its target users. The proposed system is able to infer the user and its matching Persona, and keeps shaping itself in parallel with the user's discovery of the system.","PeriodicalId":161103,"journal":{"name":"2014 14th International Conference on Hybrid Intelligent Systems","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127548185","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 : 2014-12-01DOI: 10.1109/HIS.2014.7086182
Habib Dhahri, A. Alimi, A. Abraham
In this paper, we apply the Beta Basis Function Neural Network (BBFNN) trained with cuckoo search (CS) for time series predictions. The cuckoo search algorithm optimizes the network parameters. In order to evaluate the effectiveness of the proposed method, we have carried out some experiments on four data sets: Mackey Glass, Lorenz attractor, Henon map and Box-Jenkins. We give also simulation examples to compare the effectiveness of the model with the other known methods in the literature. The results show that the CS-BBFNN model produces a better generalization performance.
{"title":"Designing of Beta Basis Function Neural Network for optimization using cuckoo search (CS)","authors":"Habib Dhahri, A. Alimi, A. Abraham","doi":"10.1109/HIS.2014.7086182","DOIUrl":"https://doi.org/10.1109/HIS.2014.7086182","url":null,"abstract":"In this paper, we apply the Beta Basis Function Neural Network (BBFNN) trained with cuckoo search (CS) for time series predictions. The cuckoo search algorithm optimizes the network parameters. In order to evaluate the effectiveness of the proposed method, we have carried out some experiments on four data sets: Mackey Glass, Lorenz attractor, Henon map and Box-Jenkins. We give also simulation examples to compare the effectiveness of the model with the other known methods in the literature. The results show that the CS-BBFNN model produces a better generalization performance.","PeriodicalId":161103,"journal":{"name":"2014 14th International Conference on Hybrid Intelligent Systems","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127586968","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 : 2014-12-01DOI: 10.1109/HIS.2014.7086173
V. Kaganov, A. Korolyov, M. Krylov, I. Mashechkin, M. Petrovskiy
Nowadays there is a growing interest to active authentication methods in security society. These methods are used for user identity validation with behavioral biometrics such as keystroke or mouse moving dynamics. In this article a new hybrid method for active authentication using keystroke dynamics is presented. This method is a combination of a new keystroke data representation model based on potential functions and machine learning algorithms based on decision trees. The proposed method is tested in static and dynamic authentication scenarios on the benchmark Si6 dataset and datasets collected by the authors. The experimental results confirm that the proposed hybrid method is applicable to reallife authentication systems.
{"title":"Hybrid method for active authentication using keystroke dynamics","authors":"V. Kaganov, A. Korolyov, M. Krylov, I. Mashechkin, M. Petrovskiy","doi":"10.1109/HIS.2014.7086173","DOIUrl":"https://doi.org/10.1109/HIS.2014.7086173","url":null,"abstract":"Nowadays there is a growing interest to active authentication methods in security society. These methods are used for user identity validation with behavioral biometrics such as keystroke or mouse moving dynamics. In this article a new hybrid method for active authentication using keystroke dynamics is presented. This method is a combination of a new keystroke data representation model based on potential functions and machine learning algorithms based on decision trees. The proposed method is tested in static and dynamic authentication scenarios on the benchmark Si6 dataset and datasets collected by the authors. The experimental results confirm that the proposed hybrid method is applicable to reallife authentication systems.","PeriodicalId":161103,"journal":{"name":"2014 14th International Conference on Hybrid Intelligent Systems","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121099082","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 : 2014-12-01DOI: 10.1109/HIS.2014.7086207
Varun Ojha, A. Abraham, V. Snás̃el
Optimization of neural network (NN) is significantly influenced by the transfer function used in its active nodes. It has been observed that the homogeneity in the activation nodes does not provide the best solution. Therefore, the customizable transfer functions whose underlying parameters are subjected to optimization were used to provide heterogeneity to NN. For experimental purposes, a meta-heuristic framework using a combined genotype representation of connection weights and transfer function parameter was used. The performance of adaptive Logistic, Tangent-hyperbolic, Gaussian and Beta functions were analyzed. Concise comparisons between different transfer function and between the NN optimization algorithms are presented. The comprehensive analysis of the results obtained over the benchmark dataset suggests that the Artificial Bee Colony with adaptive transfer function provides the best results in terms of classification accuracy over the particle swarm optimization and differential evolution algorithms.
{"title":"Simultaneous optimization of neural network weights and active nodes using metaheuristics","authors":"Varun Ojha, A. Abraham, V. Snás̃el","doi":"10.1109/HIS.2014.7086207","DOIUrl":"https://doi.org/10.1109/HIS.2014.7086207","url":null,"abstract":"Optimization of neural network (NN) is significantly influenced by the transfer function used in its active nodes. It has been observed that the homogeneity in the activation nodes does not provide the best solution. Therefore, the customizable transfer functions whose underlying parameters are subjected to optimization were used to provide heterogeneity to NN. For experimental purposes, a meta-heuristic framework using a combined genotype representation of connection weights and transfer function parameter was used. The performance of adaptive Logistic, Tangent-hyperbolic, Gaussian and Beta functions were analyzed. Concise comparisons between different transfer function and between the NN optimization algorithms are presented. The comprehensive analysis of the results obtained over the benchmark dataset suggests that the Artificial Bee Colony with adaptive transfer function provides the best results in terms of classification accuracy over the particle swarm optimization and differential evolution algorithms.","PeriodicalId":161103,"journal":{"name":"2014 14th International Conference on Hybrid Intelligent Systems","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126213384","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 : 2014-12-01DOI: 10.1109/HIS.2014.7086165
Issam Feki, A. Ammar, A. Alimi
In this paper, query sound-by-example video retrieval framework based on audio concepts is presented. First, audio stream extracted from movies in the database is set into orientation clusters using an unsupervised segmentation technique. Audio signals admit a new proposed particular pretreatment process to distinguish audio concepts. This is used for indexing the video data. Second, the query asked by the user, in sound signal form, is treated. Finally, a specific retrieval function is used to obtain video shot containing the sound of query. Objective evaluation reached 89% retrieval performance.
{"title":"Query sound-by-example video retrieval framework","authors":"Issam Feki, A. Ammar, A. Alimi","doi":"10.1109/HIS.2014.7086165","DOIUrl":"https://doi.org/10.1109/HIS.2014.7086165","url":null,"abstract":"In this paper, query sound-by-example video retrieval framework based on audio concepts is presented. First, audio stream extracted from movies in the database is set into orientation clusters using an unsupervised segmentation technique. Audio signals admit a new proposed particular pretreatment process to distinguish audio concepts. This is used for indexing the video data. Second, the query asked by the user, in sound signal form, is treated. Finally, a specific retrieval function is used to obtain video shot containing the sound of query. Objective evaluation reached 89% retrieval performance.","PeriodicalId":161103,"journal":{"name":"2014 14th International Conference on Hybrid Intelligent Systems","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125562256","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 : 2014-12-01DOI: 10.1109/HIS.2014.7086184
Oussama H. Hamid
Reinforcement learning (RL) is an algorithmic theory for learning by experience optimal action control. Two widely discussed problems within this field are the temporal credit assignment problem and the transfer of experience. The temporal credit assignment problem postulates that deciding whether an action is good or bad may not be done upon right away because of delayed rewards. The problem of transferring experience investigates the question of how experience can be generalized and transferred from a familiar context, where it was acquired, to an unfamiliar context, where it may, nevertheless, prove helpful. We propose a controller for modelling such flexibility in a context-dependent reinforcement learning paradigm. The devised controller combines two alternatives of perfect learner algorithms. In the first alternative, rewards are predicted by individual objects presented in a temporal sequence. In the second alternative, rewards are predicted on the basis of successive pairs of objects. Simulations run on both deterministic and random temporal sequences show that only in case of deterministic sequences, a previously acquired context could be retrieved. This suggests a role of temporal sequence information in the generalization and transfer of experience.
{"title":"The role of temporal statistics in the transfer of experience in context-dependent reinforcement learning","authors":"Oussama H. Hamid","doi":"10.1109/HIS.2014.7086184","DOIUrl":"https://doi.org/10.1109/HIS.2014.7086184","url":null,"abstract":"Reinforcement learning (RL) is an algorithmic theory for learning by experience optimal action control. Two widely discussed problems within this field are the temporal credit assignment problem and the transfer of experience. The temporal credit assignment problem postulates that deciding whether an action is good or bad may not be done upon right away because of delayed rewards. The problem of transferring experience investigates the question of how experience can be generalized and transferred from a familiar context, where it was acquired, to an unfamiliar context, where it may, nevertheless, prove helpful. We propose a controller for modelling such flexibility in a context-dependent reinforcement learning paradigm. The devised controller combines two alternatives of perfect learner algorithms. In the first alternative, rewards are predicted by individual objects presented in a temporal sequence. In the second alternative, rewards are predicted on the basis of successive pairs of objects. Simulations run on both deterministic and random temporal sequences show that only in case of deterministic sequences, a previously acquired context could be retrieved. This suggests a role of temporal sequence information in the generalization and transfer of experience.","PeriodicalId":161103,"journal":{"name":"2014 14th International Conference on Hybrid Intelligent Systems","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127637283","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：