Pub Date : 2009-05-15DOI: 10.1109/ESDIS.2009.4938997
J. J. Rubio, Carlos Aviles, R. Coello, Jose Francisco Cruz, Hector Rivero
In this paper, the signals of two eye movements (up an down) where taken with a MINDSET MS-100 system. A modified adaline, the multilayer neural network and the radial basis function networks where compared for pattern recognition of the two eye movements, giving that the modified adaline and the multilayer neural networks have the best behavior.
{"title":"Pattern recognition of eye movements","authors":"J. J. Rubio, Carlos Aviles, R. Coello, Jose Francisco Cruz, Hector Rivero","doi":"10.1109/ESDIS.2009.4938997","DOIUrl":"https://doi.org/10.1109/ESDIS.2009.4938997","url":null,"abstract":"In this paper, the signals of two eye movements (up an down) where taken with a MINDSET MS-100 system. A modified adaline, the multilayer neural network and the radial basis function networks where compared for pattern recognition of the two eye movements, giving that the modified adaline and the multilayer neural networks have the best behavior.","PeriodicalId":257215,"journal":{"name":"2009 IEEE Workshop on Evolving and Self-Developing Intelligent Systems","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123103323","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 : 2009-05-15DOI: 10.1109/ESDIS.2009.4938995
P. Kadlec, B. Gabrys
In this work we present an instance of an architecture for the development of robust evolving predictive models. The architecture provides a conceptual framework for the development of such models while at the same time it provides mechanisms for the minimisation of effort needed for the development and maintenance of the models. These mechanisms deal with the model and parameter selection, model training, validation and adaptation. Another challenge for the proposed instance is to deal with an industrial data set containing several issues like missing data, outliers, drifting data, etc. This fact calls for high robustness of the deployed models. The success of the models lays in the goal oriented application of several concepts like ensemble building, local learning, parameter cross-validation which are provided by the architecture and exploited by the discussed instance.
{"title":"Evolving on-line prediction model dealing with industrial data sets","authors":"P. Kadlec, B. Gabrys","doi":"10.1109/ESDIS.2009.4938995","DOIUrl":"https://doi.org/10.1109/ESDIS.2009.4938995","url":null,"abstract":"In this work we present an instance of an architecture for the development of robust evolving predictive models. The architecture provides a conceptual framework for the development of such models while at the same time it provides mechanisms for the minimisation of effort needed for the development and maintenance of the models. These mechanisms deal with the model and parameter selection, model training, validation and adaptation. Another challenge for the proposed instance is to deal with an industrial data set containing several issues like missing data, outliers, drifting data, etc. This fact calls for high robustness of the deployed models. The success of the models lays in the goal oriented application of several concepts like ensemble building, local learning, parameter cross-validation which are provided by the architecture and exploited by the discussed instance.","PeriodicalId":257215,"journal":{"name":"2009 IEEE Workshop on Evolving and Self-Developing Intelligent Systems","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131585260","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 : 2009-05-15DOI: 10.1109/ESDIS.2009.4938992
D. Leite, P. Costa, F. Gomide
This paper introduces a granular, interval-based evolving modeling (IBeM) approach to develop system models from a stream of data. IBeM is an evolving rule-based modeling scheme that gradually adapts its structure (information granules and rule base) and rules antecedent and consequent parameters from data (inductive learning). Its main purpose is continuous learning, self-organization, and adaptation to unknown environments. The IBeM approach develops global model of a system using a fast, one-pass learning algorithm, and modest memory requirements. To illustrate the effectiveness of the approach, the paper considers actual time series forecasting applications concerning electricity load and stream flow forecasting.
{"title":"Interval-based evolving modeling","authors":"D. Leite, P. Costa, F. Gomide","doi":"10.1109/ESDIS.2009.4938992","DOIUrl":"https://doi.org/10.1109/ESDIS.2009.4938992","url":null,"abstract":"This paper introduces a granular, interval-based evolving modeling (IBeM) approach to develop system models from a stream of data. IBeM is an evolving rule-based modeling scheme that gradually adapts its structure (information granules and rule base) and rules antecedent and consequent parameters from data (inductive learning). Its main purpose is continuous learning, self-organization, and adaptation to unknown environments. The IBeM approach develops global model of a system using a fast, one-pass learning algorithm, and modest memory requirements. To illustrate the effectiveness of the approach, the paper considers actual time series forecasting applications concerning electricity load and stream flow forecasting.","PeriodicalId":257215,"journal":{"name":"2009 IEEE Workshop on Evolving and Self-Developing Intelligent Systems","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132779455","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 : 2009-05-15DOI: 10.1109/ESDIS.2009.4938996
A. Bouchachia
The present paper presents an incremental fuzzy rule based system for classification purposes. Relying on fuzzy min-max neural networks, the present paper shows how fuzzy rules can be continuously online generated to meet the requirements of non-stationary dynamic environments. Simulation results are reported to show the effectiveness of the proposed approach.
{"title":"Incremental induction of fuzzy classification rules","authors":"A. Bouchachia","doi":"10.1109/ESDIS.2009.4938996","DOIUrl":"https://doi.org/10.1109/ESDIS.2009.4938996","url":null,"abstract":"The present paper presents an incremental fuzzy rule based system for classification purposes. Relying on fuzzy min-max neural networks, the present paper shows how fuzzy rules can be continuously online generated to meet the requirements of non-stationary dynamic environments. Simulation results are reported to show the effectiveness of the proposed approach.","PeriodicalId":257215,"journal":{"name":"2009 IEEE Workshop on Evolving and Self-Developing Intelligent Systems","volume":"103 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116624789","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 : 2009-05-15DOI: 10.1109/ESDIS.2009.4938998
M. Gongora, M. C. Rodas
This paper presents the application of research in evolutionary artificial developmental systems to the study of complex and organized asymmetry development in embryology systems, mainly as a tool to understand and analyze developmental processes in biology. LRA is presented as an organized level of developmental stage in reference to the 2 basic types of asymmetry presented by all developing embryos (Anterior-Posterior and Dorsal-Ventral).
{"title":"Analysis of organized asymmetry development using artificial cellular differentiation models","authors":"M. Gongora, M. C. Rodas","doi":"10.1109/ESDIS.2009.4938998","DOIUrl":"https://doi.org/10.1109/ESDIS.2009.4938998","url":null,"abstract":"This paper presents the application of research in evolutionary artificial developmental systems to the study of complex and organized asymmetry development in embryology systems, mainly as a tool to understand and analyze developmental processes in biology. LRA is presented as an organized level of developmental stage in reference to the 2 basic types of asymmetry presented by all developing embryos (Anterior-Posterior and Dorsal-Ventral).","PeriodicalId":257215,"journal":{"name":"2009 IEEE Workshop on Evolving and Self-Developing Intelligent Systems","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124645479","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 : 2009-05-15DOI: 10.1109/ESDIS.2009.4938994
J. A. Iglesias, P. Angelov, Agapito Ledezma, A. Sanchis
Knowledge about computer users is very beneficial for assisting them, predicting their future actions or detecting masqueraders. In this paper, a new approach for creating and recognizing automatically the behaviour profile of a computer user is presented. In this case, a computer user behaviour is represented as the sequence of the commands (s)he types during her/his work. This sequence is transformed into a distribution of relevant subsequences of commands in order to find out a profile that defines its behaviour. Also, because of a user profile is not necessarily fixed but rather it evolves/changes, we propose an evolving method to keep up to date the created profiles using an Evolving Systems approach. In this paper we combine the evolving classifier with a trie-based user profiling to obtain a powerful self-learning on-line scheme. We also develop further the recursive formula of the potential of a data point to become a cluster centre using cosine distance which is provided in the Appendix. The novel approach proposed in this paper can be applicable to any problem of dynamic/evolving user behaviour modelling where it can be represented as a sequence of actions and events. It has been evaluated on several real data streams.
{"title":"Modelling evolving user behaviours","authors":"J. A. Iglesias, P. Angelov, Agapito Ledezma, A. Sanchis","doi":"10.1109/ESDIS.2009.4938994","DOIUrl":"https://doi.org/10.1109/ESDIS.2009.4938994","url":null,"abstract":"Knowledge about computer users is very beneficial for assisting them, predicting their future actions or detecting masqueraders. In this paper, a new approach for creating and recognizing automatically the behaviour profile of a computer user is presented. In this case, a computer user behaviour is represented as the sequence of the commands (s)he types during her/his work. This sequence is transformed into a distribution of relevant subsequences of commands in order to find out a profile that defines its behaviour. Also, because of a user profile is not necessarily fixed but rather it evolves/changes, we propose an evolving method to keep up to date the created profiles using an Evolving Systems approach. In this paper we combine the evolving classifier with a trie-based user profiling to obtain a powerful self-learning on-line scheme. We also develop further the recursive formula of the potential of a data point to become a cluster centre using cosine distance which is provided in the Appendix. The novel approach proposed in this paper can be applicable to any problem of dynamic/evolving user behaviour modelling where it can be represented as a sequence of actions and events. It has been evaluated on several real data streams.","PeriodicalId":257215,"journal":{"name":"2009 IEEE Workshop on Evolving and Self-Developing Intelligent Systems","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132517135","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 : 2009-05-15DOI: 10.1109/ESDIS.2009.4938999
T. Smilkstein, K. Tati, Parashar Barve, M. Hai, Kittisak Sajjapongse, Durgesh K. Sharma
We have developed a general purpose evolutionary algorithm testbed (GPeat) that allows evolutionary algorithm designers to quickly and with minimal hardware knowledge move their algorithms into hardware. A user programs the testbed through a graphical user interface (GUI) that lets the user choose system parameters such as types and combinations of crossovers and mutations, initial population descriptions, fitness function rules, criteria for selection and elitism rates. A variety of sensors or computer connections can be made to the testbed so that both intrinsic and extrinsic runs can be carried out. Outputs of the testbed can likewise be computer or device directed. Use of the GUI requires minimal knowledge of hardware and connecting sensors and output devices to the board requires only the ability to identify basic device characteristics (i.e. voltage or current output, analog or digital output). In this first version, sensor inputs, fitness/chromosome value pairs, generated initial values, selected outputs are dumped to a file on the computer for analysis. New evolutionary algorithm specific hardware structures have also been developed which can provide faster run times than direct FPGA implementations. This tool will allow quick prototyping for those wanting to move their algorithms from the computer to the real world, the option to use the hardware as a debugging tool or as the final embedded, portable evolutionary algorithm hardware system.
{"title":"An evolutionary algorithm testbed for quick implementation of algorithms in hardware","authors":"T. Smilkstein, K. Tati, Parashar Barve, M. Hai, Kittisak Sajjapongse, Durgesh K. Sharma","doi":"10.1109/ESDIS.2009.4938999","DOIUrl":"https://doi.org/10.1109/ESDIS.2009.4938999","url":null,"abstract":"We have developed a general purpose evolutionary algorithm testbed (GPeat) that allows evolutionary algorithm designers to quickly and with minimal hardware knowledge move their algorithms into hardware. A user programs the testbed through a graphical user interface (GUI) that lets the user choose system parameters such as types and combinations of crossovers and mutations, initial population descriptions, fitness function rules, criteria for selection and elitism rates. A variety of sensors or computer connections can be made to the testbed so that both intrinsic and extrinsic runs can be carried out. Outputs of the testbed can likewise be computer or device directed. Use of the GUI requires minimal knowledge of hardware and connecting sensors and output devices to the board requires only the ability to identify basic device characteristics (i.e. voltage or current output, analog or digital output). In this first version, sensor inputs, fitness/chromosome value pairs, generated initial values, selected outputs are dumped to a file on the computer for analysis. New evolutionary algorithm specific hardware structures have also been developed which can provide faster run times than direct FPGA implementations. This tool will allow quick prototyping for those wanting to move their algorithms from the computer to the real world, the option to use the hardware as a debugging tool or as the final embedded, portable evolutionary algorithm hardware system.","PeriodicalId":257215,"journal":{"name":"2009 IEEE Workshop on Evolving and Self-Developing Intelligent Systems","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131824992","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 : 2009-05-15DOI: 10.1109/ESDIS.2009.4938993
J. J. R. Avila, Jaime Pacheco Martinez, A. F. Ramírez
In this research, we propose an evolving neuro-fuzzy recurrent network (ENFRN). The network is capable to perceive the change in the actual system and adapt (self organize) itself to the new situation. The network generates a new hidden neuron if the smallest distance between the new data and all the existing hidden neurons (the winner neuron) is more than a given radius. We propose a new pruning algorithm based on the density. Density is the number of times each hidden neuron is used. If the value of the smallest density (the looser neuron) is smaller to a specified umbral, this neuron is pruned. We use a modified least square algorithm to train the parameters of the network. Structure and parameters learning are updated at the same time. The major contribution of this research is: we present the stability of the algorithm of the evolving neuro-fuzzy reccurrent network proposed. Two simulations give the effectiveness of the suggested algorithm.
{"title":"An evolving neuro-fuzzy recurrent network","authors":"J. J. R. Avila, Jaime Pacheco Martinez, A. F. Ramírez","doi":"10.1109/ESDIS.2009.4938993","DOIUrl":"https://doi.org/10.1109/ESDIS.2009.4938993","url":null,"abstract":"In this research, we propose an evolving neuro-fuzzy recurrent network (ENFRN). The network is capable to perceive the change in the actual system and adapt (self organize) itself to the new situation. The network generates a new hidden neuron if the smallest distance between the new data and all the existing hidden neurons (the winner neuron) is more than a given radius. We propose a new pruning algorithm based on the density. Density is the number of times each hidden neuron is used. If the value of the smallest density (the looser neuron) is smaller to a specified umbral, this neuron is pruned. We use a modified least square algorithm to train the parameters of the network. Structure and parameters learning are updated at the same time. The major contribution of this research is: we present the stability of the algorithm of the evolving neuro-fuzzy reccurrent network proposed. Two simulations give the effectiveness of the suggested algorithm.","PeriodicalId":257215,"journal":{"name":"2009 IEEE Workshop on Evolving and Self-Developing Intelligent Systems","volume":"38 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131662412","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 : 2009-05-15DOI: 10.1109/ESDIS.2009.4939000
Huajin Tang, C. H. Tan, Kay Chen Tan, A. Tay
This paper presents examines the design of controllers that is computationally efficient yet demonstrates highly competitive performance for a real time simulated car racing game. Algorithms that require large amount of computational resources are impractical for fast paced and real time games (i.e. racing games, sports simulators, first person shooters and real time strategy games). This paper examines the design of two computationally efficient approaches, neural networks and behaviour based intelligence, in the context of a real time car racing game. Both approaches are optimized using evolutionary strategies. The behaviour based approach was found to obtain a higher fitness value yet being more computationally efficient. The design approaches can also be applied to real-time face animation which involves data-intensive computations.
{"title":"Neural network versus behavior based approach in simulated car racing game","authors":"Huajin Tang, C. H. Tan, Kay Chen Tan, A. Tay","doi":"10.1109/ESDIS.2009.4939000","DOIUrl":"https://doi.org/10.1109/ESDIS.2009.4939000","url":null,"abstract":"This paper presents examines the design of controllers that is computationally efficient yet demonstrates highly competitive performance for a real time simulated car racing game. Algorithms that require large amount of computational resources are impractical for fast paced and real time games (i.e. racing games, sports simulators, first person shooters and real time strategy games). This paper examines the design of two computationally efficient approaches, neural networks and behaviour based intelligence, in the context of a real time car racing game. Both approaches are optimized using evolutionary strategies. The behaviour based approach was found to obtain a higher fitness value yet being more computationally efficient. The design approaches can also be applied to real-time face animation which involves data-intensive computations.","PeriodicalId":257215,"journal":{"name":"2009 IEEE Workshop on Evolving and Self-Developing Intelligent Systems","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2009-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116734514","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：