Pub Date : 2005-12-15DOI: 10.1109/CIMA.2005.1662358
A. Atar, O. Ersoy, L. Ozyilmaz
It is necessary to know both the primary and secondary structure of proteins in order to predict their biological functions. Neural networks are effective for secondary structure prediction of proteins. In this study, the self-organizing map (SOM) algorithm, and the self-organizing global ranking (SOGR) algorithm were investigated with different window sizes of amino acid sequences to predict the protein secondary structure from the protein primary structure. In this study, all of the data were obtained from PDB (protein data bank). Then, the letter data were converted to numerical data and processed with ANNs. 17 different types of data with a number of sliding window lengths were used. In general, results were very satisfactory, and the SOGR had the highest testing accuracies and faster speed of learning
为了预测蛋白质的生物学功能，有必要了解蛋白质的一级和二级结构。神经网络是一种有效的蛋白质二级结构预测方法。本文研究了自组织映射(SOM)算法和自组织全局排序(SOGR)算法在不同的氨基酸序列窗口大小下，从蛋白质一级结构预测蛋白质二级结构。在本研究中，所有数据均来自PDB (protein data bank)。然后，将字母数据转换为数字数据，并用人工神经网络进行处理。使用了17种不同类型的数据和若干滑动窗口长度。总的来说，结果非常令人满意，SOGR具有最高的测试精度和更快的学习速度
{"title":"Prediction of protein secondary structure by SOM and SOGR algorithms","authors":"A. Atar, O. Ersoy, L. Ozyilmaz","doi":"10.1109/CIMA.2005.1662358","DOIUrl":"https://doi.org/10.1109/CIMA.2005.1662358","url":null,"abstract":"It is necessary to know both the primary and secondary structure of proteins in order to predict their biological functions. Neural networks are effective for secondary structure prediction of proteins. In this study, the self-organizing map (SOM) algorithm, and the self-organizing global ranking (SOGR) algorithm were investigated with different window sizes of amino acid sequences to predict the protein secondary structure from the protein primary structure. In this study, all of the data were obtained from PDB (protein data bank). Then, the letter data were converted to numerical data and processed with ANNs. 17 different types of data with a number of sliding window lengths were used. In general, results were very satisfactory, and the SOGR had the highest testing accuracies and faster speed of learning","PeriodicalId":306045,"journal":{"name":"2005 ICSC Congress on Computational Intelligence Methods and Applications","volume":"26 7","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120823015","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 : 2005-12-15DOI: 10.1109/CIMA.2005.1662345
S. Zhou, Zeng-qi Sun
A `weak' learning algorithm that performs just slightly better than random guessing can be `boosted' into an arbitrarily accurate `strong' learning algorithm by Schapire, R.E., (1990), Inspired from the `ensemble method' idea, the paper proposes a novel conceptive model of EDA ensemble: a collection of EDAs are used to optimize the same problem, during the evolution process information interaction happens among EDAs, and at last optimum solutions can be obtained more likely than a single `strong' EDA. As an instance, PBIL ensemble model is designed in details. Every PBIL serves as a component in PBIL ensemble and cooperate with others to efficiently accomplish an optimization process. Experiments on knapsack problems and function optimization problems show that PBIL ensemble exhibits better performance than simple GA and PBIL. And amazingly, to the GA-hard problem, e.g. 4-order fully deceptive problem, PBIL ensemble can achieve the optimal solution almost all the time
{"title":"PBIL ensemble: many better than one","authors":"S. Zhou, Zeng-qi Sun","doi":"10.1109/CIMA.2005.1662345","DOIUrl":"https://doi.org/10.1109/CIMA.2005.1662345","url":null,"abstract":"A `weak' learning algorithm that performs just slightly better than random guessing can be `boosted' into an arbitrarily accurate `strong' learning algorithm by Schapire, R.E., (1990), Inspired from the `ensemble method' idea, the paper proposes a novel conceptive model of EDA ensemble: a collection of EDAs are used to optimize the same problem, during the evolution process information interaction happens among EDAs, and at last optimum solutions can be obtained more likely than a single `strong' EDA. As an instance, PBIL ensemble model is designed in details. Every PBIL serves as a component in PBIL ensemble and cooperate with others to efficiently accomplish an optimization process. Experiments on knapsack problems and function optimization problems show that PBIL ensemble exhibits better performance than simple GA and PBIL. And amazingly, to the GA-hard problem, e.g. 4-order fully deceptive problem, PBIL ensemble can achieve the optimal solution almost all the time","PeriodicalId":306045,"journal":{"name":"2005 ICSC Congress on Computational Intelligence Methods and Applications","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122400574","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 : 2005-12-15DOI: 10.1109/CIMA.2005.1662318
Hoi-Ming Chi, O. Ersoy, H. Moskowitz
A novel feature selection algorithm that combines the ideas of linear support vector machines (SVMs) and random probes is proposed. A random probe is first artificially generated from a Gaussian distribution and appended to the data set as an extra input variable. Next, a standard 2-norm or 1-norm linear support vector machine is trained using this new data set. Each coefficient, or weight, in a linear SVM is compared to that of the random probe feature. Under several statistical assumptions, the probability of each input feature being more relevant than the random probe can be computed easily. The proposed feature selection method is intuitive to use in real-world problems, and it automatically determines the optimal number of features needed. It can also be extended to selecting significant interaction and/or quadratic terms in a 2nd-order polynomial representation
{"title":"Feature selection using random probes and linear support vector machines","authors":"Hoi-Ming Chi, O. Ersoy, H. Moskowitz","doi":"10.1109/CIMA.2005.1662318","DOIUrl":"https://doi.org/10.1109/CIMA.2005.1662318","url":null,"abstract":"A novel feature selection algorithm that combines the ideas of linear support vector machines (SVMs) and random probes is proposed. A random probe is first artificially generated from a Gaussian distribution and appended to the data set as an extra input variable. Next, a standard 2-norm or 1-norm linear support vector machine is trained using this new data set. Each coefficient, or weight, in a linear SVM is compared to that of the random probe feature. Under several statistical assumptions, the probability of each input feature being more relevant than the random probe can be computed easily. The proposed feature selection method is intuitive to use in real-world problems, and it automatically determines the optimal number of features needed. It can also be extended to selecting significant interaction and/or quadratic terms in a 2nd-order polynomial representation","PeriodicalId":306045,"journal":{"name":"2005 ICSC Congress on Computational Intelligence Methods and Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126034590","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 : 2005-12-15DOI: 10.1109/CIMA.2005.1662365
V. Milea, J. van den Berg
We propose a new ants algorithm (AA) for predicting bankruptcy of firms. Inspired by an AA for solving the travelling salesman problem, we present the choices made in order to transform it into the AA for bankruptcy prediction. We sketch the experimental set-up of our simulations and the origin of the data sets used. Next we show the most important results obtained. It turns out that in some cases our AA outperforms the classical multiple discriminant analysis method as introduced by Altman in his seminal paper (Altmann, 1968). We finalize by offering a discussion and a few conclusions
{"title":"An ants algorithm for bankruptcy prediction","authors":"V. Milea, J. van den Berg","doi":"10.1109/CIMA.2005.1662365","DOIUrl":"https://doi.org/10.1109/CIMA.2005.1662365","url":null,"abstract":"We propose a new ants algorithm (AA) for predicting bankruptcy of firms. Inspired by an AA for solving the travelling salesman problem, we present the choices made in order to transform it into the AA for bankruptcy prediction. We sketch the experimental set-up of our simulations and the origin of the data sets used. Next we show the most important results obtained. It turns out that in some cases our AA outperforms the classical multiple discriminant analysis method as introduced by Altman in his seminal paper (Altmann, 1968). We finalize by offering a discussion and a few conclusions","PeriodicalId":306045,"journal":{"name":"2005 ICSC Congress on Computational Intelligence Methods and Applications","volume":"199 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127137242","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 : 2005-12-15DOI: 10.1109/CIMA.2005.1662298
B. Tutmez, A. Tercan
The estimation of reserves is directly related to the financial investments. In this study, fuzzy approach was presented for estimating ore grades which is the main part of reserve estimation. Data driven fuzzy modelling technique is applied to estimate the grades in a trona (natural soda) deposit. Performance evaluations of the fuzzy model showed that fuzzy modelling is a useful tool for the estimation of ore parameters
{"title":"Estimation of trona grade by fuzzy modelling","authors":"B. Tutmez, A. Tercan","doi":"10.1109/CIMA.2005.1662298","DOIUrl":"https://doi.org/10.1109/CIMA.2005.1662298","url":null,"abstract":"The estimation of reserves is directly related to the financial investments. In this study, fuzzy approach was presented for estimating ore grades which is the main part of reserve estimation. Data driven fuzzy modelling technique is applied to estimate the grades in a trona (natural soda) deposit. Performance evaluations of the fuzzy model showed that fuzzy modelling is a useful tool for the estimation of ore parameters","PeriodicalId":306045,"journal":{"name":"2005 ICSC Congress on Computational Intelligence Methods and Applications","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127040263","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 : 2005-12-15DOI: 10.1109/CIMA.2005.1662301
V. Hue, J. Fourquet, P. Chiron
This paper deals with modelling of human movement in order to provide an analysis and simulation tool which could be used for workspace design. We are interested in systematic generation of human motion and postures that correspond to working tasks defined by grasps or, more generally, by locations of particular frames linked to hands, head or another body of interest. These tasks are reaching - point to point - tasks or tasks with imposed Cartesian path. So the problem is: how to control global posture when frames attached to some bodies are imposed? In the general case, the constraints imposed by the task are equality constraints imposed by kinematics on one side, and obstacles or articular bounds defining inequality constraints on another side. When there is less equality constraints than independent control variables, then the problem is redundant and many solutions realize the constraints. Since the goal is to provide a realistic simulation in order to predict real human motion, redundancy has to be used to obtain realistic movements, to avoid collisions with 3D environment and to take into account articular limitations
{"title":"Simulation and modelling of virtual operators for workspace design","authors":"V. Hue, J. Fourquet, P. Chiron","doi":"10.1109/CIMA.2005.1662301","DOIUrl":"https://doi.org/10.1109/CIMA.2005.1662301","url":null,"abstract":"This paper deals with modelling of human movement in order to provide an analysis and simulation tool which could be used for workspace design. We are interested in systematic generation of human motion and postures that correspond to working tasks defined by grasps or, more generally, by locations of particular frames linked to hands, head or another body of interest. These tasks are reaching - point to point - tasks or tasks with imposed Cartesian path. So the problem is: how to control global posture when frames attached to some bodies are imposed? In the general case, the constraints imposed by the task are equality constraints imposed by kinematics on one side, and obstacles or articular bounds defining inequality constraints on another side. When there is less equality constraints than independent control variables, then the problem is redundant and many solutions realize the constraints. Since the goal is to provide a realistic simulation in order to predict real human motion, redundancy has to be used to obtain realistic movements, to avoid collisions with 3D environment and to take into account articular limitations","PeriodicalId":306045,"journal":{"name":"2005 ICSC Congress on Computational Intelligence Methods and Applications","volume":"78 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132553117","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 : 2005-12-15DOI: 10.1109/CIMA.2005.1662366
D. Grzechca, J. Rutkowski
This paper presents a new concept to analog fault diagnosis. Problem of distinguishing between healthy or faulty analog circuit has always been very complicated. The most common approach based on pattern recognition, especially on mean square error measure, can not distinguish all faulty circuits from the healthy one. Normally, the dictionary has to include thousands of patterns and even then, the level of fault detection is not satisfactory. A neural network classifier has been proposed to solve the problem. Its generalization ability allows to reduce the dictionary size significantly. This paper shows how to create a neural dictionary for fault location. Moreover, at the first stage of classification, the fuzzy logic is utilized to transform a measurement vector into a zero-one range. The information from the circuit under test (CUT) has to be as high as it is possible but at the same time the stimuli has to be as simple as possible. The most common AC and DC tests don't give the best solution. Therefore, the time domain testing with pulse stimuli has been utilized. This paper presents a new concept to analog fault diagnosis
{"title":"Use of neuro-fuzzy system to time domain electronic circuits fault diagnosis","authors":"D. Grzechca, J. Rutkowski","doi":"10.1109/CIMA.2005.1662366","DOIUrl":"https://doi.org/10.1109/CIMA.2005.1662366","url":null,"abstract":"This paper presents a new concept to analog fault diagnosis. Problem of distinguishing between healthy or faulty analog circuit has always been very complicated. The most common approach based on pattern recognition, especially on mean square error measure, can not distinguish all faulty circuits from the healthy one. Normally, the dictionary has to include thousands of patterns and even then, the level of fault detection is not satisfactory. A neural network classifier has been proposed to solve the problem. Its generalization ability allows to reduce the dictionary size significantly. This paper shows how to create a neural dictionary for fault location. Moreover, at the first stage of classification, the fuzzy logic is utilized to transform a measurement vector into a zero-one range. The information from the circuit under test (CUT) has to be as high as it is possible but at the same time the stimuli has to be as simple as possible. The most common AC and DC tests don't give the best solution. Therefore, the time domain testing with pulse stimuli has been utilized. This paper presents a new concept to analog fault diagnosis","PeriodicalId":306045,"journal":{"name":"2005 ICSC Congress on Computational Intelligence Methods and Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131233164","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 : 2005-12-15DOI: 10.1109/CIMA.2005.1662309
Rong Li, Weiguo Liu, Ruiqing Ma, Yashan Hu
Electric actuator dual-redundancy servo system consists of dual-redundancy rare earth permanent magnet (REPM) brushless DC machine (BLDCM), digital signal processor (DSP) and some periphery circuits. Here, dynamic and static states of current symmetry on dual-redundancy working is emphatically investigated. It analyses the causes of current asymmetry and possible consequences, gives a method of current symmetry and makes use of it in a real system, compares the experiment results of current symmetry and asymmetry in this system, which shows that the method is simply and practically to be reality, two redundancies currents are more obviously consistent after adding current symmetry to this system than before, the system performance is improved greatly
{"title":"Research of current symmetry on dual-redundancy","authors":"Rong Li, Weiguo Liu, Ruiqing Ma, Yashan Hu","doi":"10.1109/CIMA.2005.1662309","DOIUrl":"https://doi.org/10.1109/CIMA.2005.1662309","url":null,"abstract":"Electric actuator dual-redundancy servo system consists of dual-redundancy rare earth permanent magnet (REPM) brushless DC machine (BLDCM), digital signal processor (DSP) and some periphery circuits. Here, dynamic and static states of current symmetry on dual-redundancy working is emphatically investigated. It analyses the causes of current asymmetry and possible consequences, gives a method of current symmetry and makes use of it in a real system, compares the experiment results of current symmetry and asymmetry in this system, which shows that the method is simply and practically to be reality, two redundancies currents are more obviously consistent after adding current symmetry to this system than before, the system performance is improved greatly","PeriodicalId":306045,"journal":{"name":"2005 ICSC Congress on Computational Intelligence Methods and Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115875477","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 : 2005-12-15DOI: 10.1109/CIMA.2005.1662363
M. D. D. Gelder, U. Kaymak, K. Boer, E. Asperen
The minority game has been proposed as a simplified model for studying dynamics in financial markets. However, minority games assume synchronous trading, while most financial markets are asynchronous. In this paper, we investigate asynchronous models for the minority games within a framework for asynchronous continuous trading in financial markets. We propose four possible extensions to the synchronous game and study their effects on the market dynamics
{"title":"Minority games with asynchronous trading","authors":"M. D. D. Gelder, U. Kaymak, K. Boer, E. Asperen","doi":"10.1109/CIMA.2005.1662363","DOIUrl":"https://doi.org/10.1109/CIMA.2005.1662363","url":null,"abstract":"The minority game has been proposed as a simplified model for studying dynamics in financial markets. However, minority games assume synchronous trading, while most financial markets are asynchronous. In this paper, we investigate asynchronous models for the minority games within a framework for asynchronous continuous trading in financial markets. We propose four possible extensions to the synchronous game and study their effects on the market dynamics","PeriodicalId":306045,"journal":{"name":"2005 ICSC Congress on Computational Intelligence Methods and Applications","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116444877","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 : 2005-12-15DOI: 10.1109/CIMA.2005.1662329
S. Bhogal, K. Revett
A surprisingly large number of animal toxins target voltage sensitive ion channels. Even though there exists toxins for all four major voltage sensitive ion channels, a majority act either on sodium or potassium channels. Given a specific primary sequence, the challenge is to determine in an automated fashion whether a given substance is toxic, and what its site of action might be. Currently, there are signals such as functional dyads that are indicative of a toxin, but are not yet specific enough to allow accurate prediction of the site of action. In this paper, an automated approach for detecting whether a toxin acts on voltage-sensitive sodium versus potassium channels is presented. In addition, our consensus sequence is also able to reliably determine whether the toxin acts as a gate modifier or pore blocker (> 93% accuracy)
{"title":"Animal toxins: what features differentiate pore blockers from gate modifiers?","authors":"S. Bhogal, K. Revett","doi":"10.1109/CIMA.2005.1662329","DOIUrl":"https://doi.org/10.1109/CIMA.2005.1662329","url":null,"abstract":"A surprisingly large number of animal toxins target voltage sensitive ion channels. Even though there exists toxins for all four major voltage sensitive ion channels, a majority act either on sodium or potassium channels. Given a specific primary sequence, the challenge is to determine in an automated fashion whether a given substance is toxic, and what its site of action might be. Currently, there are signals such as functional dyads that are indicative of a toxin, but are not yet specific enough to allow accurate prediction of the site of action. In this paper, an automated approach for detecting whether a toxin acts on voltage-sensitive sodium versus potassium channels is presented. In addition, our consensus sequence is also able to reliably determine whether the toxin acts as a gate modifier or pore blocker (> 93% accuracy)","PeriodicalId":306045,"journal":{"name":"2005 ICSC Congress on Computational Intelligence Methods and Applications","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2005-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114892343","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：