International Journal of Artificial Intelligence and Soft Computing http://www.inderscience.com/browse/index.php?journalID=280&year=2011&vol=2&issue=4 Inderscience Publishers Ltd en-uk International Journal of Artificial Intelligence and Soft Computing 1755-4950 1755-4969 © 2011 Inderscience Publishers Ltd editor@inderscience.com https://www.inderscience.com/images/files/coverImgs/ijaisc_scoverijaisc.jpg http://www.inderscience.com/browse/index.php?journalID=280&year=2011&vol=2&issue=4 http://www.inderscience.com/link.php?id=42710 Industry has monitoring systems to determine the tool condition and to ensure quality. This paper presents an intelligent classification system which determines the status of cutters in a CNC milling machine. The tool states are detected through the analysis of the cutting forces drawn from the spindle motors currents. A wavelet transformation was used in order to compress the data and to optimise the classifier structure. Then a supervised SOM neural network is responsible for carrying out the classification of the signal. Achieving a reliability of 95%, the system is capable of detecting breakage and a worn cutter. Sensorless intelligent classifier of tool condition in a CNC milling machine using a SOM supervised neural network
Georgina Del Carmen Mota-Valtierra; Luis Alfonso Franco-Gasca; Gilberto Herrera-Ruiz
International Journal of Artificial Intelligence and Soft Computing, Vol. 2, No. 4 (2011) pp. 263 - 271
Industry has monitoring systems to determine the tool condition and to ensure quality. This paper presents an intelligent classification system which determines the status of cutters in a CNC milling machine. The tool states are detected through the analysis of the cutting forces drawn from the spindle motors currents. A wavelet transformation was used in order to compress the data and to optimise the classifier structure. Then a supervised SOM neural network is responsible for carrying out the classification of the signal. Achieving a reliability of 95%, the system is capable of detecting breakage and a worn cutter.

]]> 10.1504/IJAISC.2011.042710 International Journal of Artificial Intelligence and Soft Computing, Vol. 2, No. 4 (2011) pp. 263 - 271 Georgina Del Carmen Mota-Valtierra; Luis Alfonso Franco-Gasca; Gilberto Herrera-Ruiz Facultad de Ingenieria, Laboratorio de Mecatronica, Universidad Autonoma de Queretaro, Cerro de las campanas s/n, Queretaro, Qro. CP 76010, Mexico. ' CIATEQ, A.C., LabCASD, Av. del Retablo 150, Col. Fovisste, Queretaro, Qro. CP 76150, Mexico. ' Facultad de Ingenieria, Laboratorio de Mecatronica, Universidad Autonoma de Queretaro, Cerro de las campanas s/n, Queretaro, Qro. CP 76010, Mexico tool breakage wear monitoring sensorless classifiers intelligent classification tool condition monitoring CNC milling SOM neural networks supervised neural networks tool wear tool monitoring wavelet transforms. 2011-09-27T23:20:50-05:00 2 4 263 271 2011-09-27T23:20:50-05:00 http://www.inderscience.com/link.php?id=42711 We describe a preliminary investigation into learning a Chess player's style from game records. The method is based on attempting to learn features of a player's individual evaluation function using the method of temporal differences, with the aid of a conventional Chess engine architecture. Some encouraging results were obtained in learning the styles of two Chess world champions, and we report on our attempt to use the learnt styles to discriminate between the players from game records, by trying to detect who was playing white and who was playing black. We also discuss some limitations of our approach. A methodology for learning players' styles from game records
Mark Levene; Trevor Fenner
International Journal of Artificial Intelligence and Soft Computing, Vol. 2, No. 4 (2011) pp. 272 - 286
We describe a preliminary investigation into learning a Chess player's style from game records. The method is based on attempting to learn features of a player's individual evaluation function using the method of temporal differences, with the aid of a conventional Chess engine architecture. Some encouraging results were obtained in learning the styles of two Chess world champions, and we report on our attempt to use the learnt styles to discriminate between the players from game records, by trying to detect who was playing white and who was playing black. We also discuss some limitations of our approach.

]]> 10.1504/IJAISC.2011.042711 International Journal of Artificial Intelligence and Soft Computing, Vol. 2, No. 4 (2011) pp. 272 - 286 Mark Levene; Trevor Fenner Department of Computer Science and Information Systems, Birkbeck College, University of London, London WC1E 7HX, UK. ' Department of Computer Science and Information Systems, Birkbeck College, University of London, London WC1E 7HX, UK temporal difference learning evaluation function game records player styles computer chess chess players game playing gaming playing styles. 2011-09-27T23:20:50-05:00 2 4 272 286 2011-09-27T23:20:50-05:00 http://www.inderscience.com/link.php?id=42712 Precise and accurate automatic recognition of decimal numbers is essential for many applications. Motion Estimation (ME) is a basic component in any video compression technique used to account for the temporal redundancy in image sequences. Global Motions are often modelled by parametric transformations of two dimensional images. The process of estimating the motion parameters is called Global Motion Estimation (GME). In this paper, we propose a new way of using GME for the purpose of off-line machine-print decimal digit. We show that the proposed approach is able to achieve very high recognition rates. Precise and accurate decimal number recognition using Global Motion Estimation
Hussein R. Al-Zoubi; Mahmood A. Al-Khassaweneh; Amin T. Alqudah
International Journal of Artificial Intelligence and Soft Computing, Vol. 2, No. 4 (2011) pp. 287 - 301
Precise and accurate automatic recognition of decimal numbers is essential for many applications. Motion Estimation (ME) is a basic component in any video compression technique used to account for the temporal redundancy in image sequences. Global Motions are often modelled by parametric transformations of two dimensional images. The process of estimating the motion parameters is called Global Motion Estimation (GME). In this paper, we propose a new way of using GME for the purpose of off-line machine-print decimal digit. We show that the proposed approach is able to achieve very high recognition rates.

]]> 10.1504/IJAISC.2011.042712 International Journal of Artificial Intelligence and Soft Computing, Vol. 2, No. 4 (2011) pp. 287 - 301 Hussein R. Al-Zoubi; Mahmood A. Al-Khassaweneh; Amin T. Alqudah Hijjawi Faculty for Engineering Technology, Computer Engineering Department, Yarmouk University, Irbid 21163, Jordan. ' Hijjawi Faculty for Engineering Technology, Computer Engineering Department, Yarmouk University, Irbid 21163, Jordan. ' Hijjawi Faculty for Engineering Technology, Computer Engineering Department, Yarmouk University, Irbid 21163, Jordan GME global motion estimation recognition rate motion vector Levenberg-Marquardt algorithm 3-SS three-step search arti?cial intelligence decimal number recognition video compression. 2011-09-27T23:20:50-05:00 2 4 287 301 2011-09-27T23:20:50-05:00 http://www.inderscience.com/link.php?id=42717 This paper emphasised an approach for offline signature verification and identification. Two image descriptors are studied, including Pyramid Histogram of Oriented Gradients (PHOG), and a direction feature proposed in the literature. Compared with many previously proposed signature feature extraction approaches, PHOG has advantages in the extraction of discriminative information from handwriting signature images. The significance of classification framework is stressed. With the benchmarking database "Grupo de Procesado Digital de Senales" (GPDS), satisfactory performances were obtained from several classifiers. Among the classifiers compared, SVM is clearly superior, giving a False Rejection Rate (FRR) of 2.5% and a False Acceptance Rate (FAR) 2% for skillful forgery, which compares sharply with the latest published results on the same dataset. This substantiates the superiority of the proposed method. The related issue offline signature recognition is also investigated based on the same approach, with an accuracy of 99% on the GPDS data from SVM classification. Offline signature verification and identification by hybrid features and Support Vector Machine
Bailing Zhang
International Journal of Artificial Intelligence and Soft Computing, Vol. 2, No. 4 (2011) pp. 302 - 320
This paper emphasised an approach for offline signature verification and identification. Two image descriptors are studied, including Pyramid Histogram of Oriented Gradients (PHOG), and a direction feature proposed in the literature. Compared with many previously proposed signature feature extraction approaches, PHOG has advantages in the extraction of discriminative information from handwriting signature images. The significance of classification framework is stressed. With the benchmarking database "Grupo de Procesado Digital de Senales" (GPDS), satisfactory performances were obtained from several classifiers. Among the classifiers compared, SVM is clearly superior, giving a False Rejection Rate (FRR) of 2.5% and a False Acceptance Rate (FAR) 2% for skillful forgery, which compares sharply with the latest published results on the same dataset. This substantiates the superiority of the proposed method. The related issue offline signature recognition is also investigated based on the same approach, with an accuracy of 99% on the GPDS data from SVM classification.

]]> 10.1504/IJAISC.2011.042717 International Journal of Artificial Intelligence and Soft Computing, Vol. 2, No. 4 (2011) pp. 302 - 320 Bailing Zhang Department of Computer Science and Software Engineering, Xi'an Jiaotong-Liverpool University, Suzhou, 215123, China offline signature verification offline signature recognition classification PHOG pyramid histogram of oriented gradients direction features SVM support vector machines. 2011-09-27T23:20:50-05:00 2 4 302 320 2011-09-27T23:20:50-05:00 http://www.inderscience.com/link.php?id=42713 An improved Optical Backpropagation (OBP) algorithm for training single hidden layer feedforward neural network with third term is proposed. The major limitations of backpropagation algorithm are the local minima problem and the slow rate of convergence. To solve these problems, we have proposed an algorithm by introducing a third term with optical backpropagation (OBPWT). This method has been applied to the multilayer neural network to improve the efficiency in terms of convergence speed. In the proposed algorithm, a non-linear function on the error term is introduced before applying the backpropagation phase. This error term is used along with a third term in the weight updation rule. We have shown how the new proposed algorithm drastically accelerates the training convergence at the same time maintaining the neural network’s performance. The effectiveness of the proposed algorithm has been shown by testing five benchmark problems. The simulation results show that the proposed algorithm is capable of speeding up the learning and hence the rate of convergence. An improved three-term optical backpropagation algorithm
M. Sornam; P. Thangavel
International Journal of Artificial Intelligence and Soft Computing, Vol. 2, No. 4 (2011) pp. 321 - 333
An improved Optical Backpropagation (OBP) algorithm for training single hidden layer feedforward neural network with third term is proposed. The major limitations of backpropagation algorithm are the local minima problem and the slow rate of convergence. To solve these problems, we have proposed an algorithm by introducing a third term with optical backpropagation (OBPWT). This method has been applied to the multilayer neural network to improve the efficiency in terms of convergence speed. In the proposed algorithm, a non-linear function on the error term is introduced before applying the backpropagation phase. This error term is used along with a third term in the weight updation rule. We have shown how the new proposed algorithm drastically accelerates the training convergence at the same time maintaining the neural network’s performance. The effectiveness of the proposed algorithm has been shown by testing five benchmark problems. The simulation results show that the proposed algorithm is capable of speeding up the learning and hence the rate of convergence.

]]> 10.1504/IJAISC.2011.042713 International Journal of Artificial Intelligence and Soft Computing, Vol. 2, No. 4 (2011) pp. 321 - 333 M. Sornam; P. Thangavel Department of Computer Science, University of Madras, Chepauk, Chennai 600005, Tamil Nadu, India. ' Department of Computer Science, University of Madras, Chepauk, Chennai 600005, Tamil Nadu, India optical backpropagation multilayer neural networks weight updation nonlinear function third term convergence speed. 2011-09-27T23:20:50-05:00 2 4 321 333 2011-09-27T23:20:50-05:00 http://www.inderscience.com/link.php?id=42714 Some resolution strategies, such as SLD-resolution, are such that a derivation may be infinite even on a logic program that has a finite Herbrand universe. This paper introduces GOPT-resolution, a new deduction strategy for deriving solutions from a set of rules that improves on previous methods by preventing derivations that have infinite recursion. The paper outlines the process behind the development of GOPT-resolution based on PT-resolution. GOPT-resolution is then developed by distinguishing between goal-relevant and goal-irrelevant P-domains. GOPT Resolution
Fei Liu; John Roddick
International Journal of Artificial Intelligence and Soft Computing, Vol. 2, No. 4 (2011) pp. 334 - 352
Some resolution strategies, such as SLD-resolution, are such that a derivation may be infinite even on a logic program that has a finite Herbrand universe. This paper introduces GOPT-resolution, a new deduction strategy for deriving solutions from a set of rules that improves on previous methods by preventing derivations that have infinite recursion. The paper outlines the process behind the development of GOPT-resolution based on PT-resolution. GOPT-resolution is then developed by distinguishing between goal-relevant and goal-irrelevant P-domains.

]]> 10.1504/IJAISC.2011.042714 International Journal of Artificial Intelligence and Soft Computing, Vol. 2, No. 4 (2011) pp. 334 - 352 Fei Liu; John Roddick Department of Computer Science and Computer Engineering, La Trobe University, Bundoora, Victoria 3086, Australia ' School of Computer Science, Engineering and Mathematics, Flinders University, P.O. Box 2100, Adelaide 5001, South Australia GOPT resolution logic programming resolution strategies set theory. 2011-09-27T23:20:50-05:00 2 4 334 352 2011-09-27T23:20:50-05:00 http://www.inderscience.com/link.php?id=42716 This paper discusses the bio-inspired algorithm of the Particle Swarm Optimisation (PSO) for a wheeled robot's displacement. PSO was selected because its flexibility and its tempting results. An omnidirectional wheeled robot was simulated on a flat environment with two tasks: 'Reach a goal' or 'collect balls'. This paper checks on the performance of PSO for the displacement studied. In the first case, we discussed the variation of execution time compared to the particles' and the neighbours' number. In the second one, we studied the change in the path's length compared to execution time depending on the particles' and balls' number. Bio-inspired algorithm for wheeled robot's navigation
Nada Kherici; Yamina Mohamed Ben Ali
International Journal of Artificial Intelligence and Soft Computing, Vol. 2, No. 4 (2011) pp. 353 - 366
This paper discusses the bio-inspired algorithm of the Particle Swarm Optimisation (PSO) for a wheeled robot's displacement. PSO was selected because its flexibility and its tempting results. An omnidirectional wheeled robot was simulated on a flat environment with two tasks: 'Reach a goal' or 'collect balls'. This paper checks on the performance of PSO for the displacement studied. In the first case, we discussed the variation of execution time compared to the particles' and the neighbours' number. In the second one, we studied the change in the path's length compared to execution time depending on the particles' and balls' number.

]]> 10.1504/IJAISC.2011.042716 International Journal of Artificial Intelligence and Soft Computing, Vol. 2, No. 4 (2011) pp. 353 - 366 Nada Kherici; Yamina Mohamed Ben Ali LRI (Laboratory of Computer Research), Computer Science Department, University of Badji Mokhtar, P.O. Box No. 12, El Hadjar 23200, Algeria. ' LRI (Laboratory of Computer Research), Computer Science Department, University of Badji Mokhtar, P.O. Box No. 12, El Hadjar 23200, Algeria autonomous navigation bio-inspired metaheuristics evolutionary robotics mobile robots wheeled robots PSO particles swarm optimisation artificial intelligence displacement robot navigation. 2011-09-27T23:20:50-05:00 2 4 353 366 2011-09-27T23:20:50-05:00 http://www.inderscience.com/link.php?id=42715 A full customised case-oriented Multi-Layered Recurrent Neural Network (MLRNN) has been proposed to predict the Capacity of Crude Oil Distillation in OPEC Member Countries. Recurrent neural networks use feedback connections and have the potential to represent certain computational structures in a more parsimonious fashion. Moreover, a cluster based training procedure, in which proper opportunity achieves for network to sense complicated nonlinear relations in data, has been supplied. The results of proposed MLRNN were promising in comparison with the results of a Multi-Layered Feed-Forward Neural Network (MLFFNN) on the aforementioned case study. Application of multi-layer recurrent neural network in chaotic time series prediction: a real case study of crude oil distillation capacity
Kaveh Khalili-Damghani; Soheil Sadi-Nezhad
International Journal of Artificial Intelligence and Soft Computing, Vol. 2, No. 4 (2011) pp. 367 - 380
A full customised case-oriented Multi-Layered Recurrent Neural Network (MLRNN) has been proposed to predict the Capacity of Crude Oil Distillation in OPEC Member Countries. Recurrent neural networks use feedback connections and have the potential to represent certain computational structures in a more parsimonious fashion. Moreover, a cluster based training procedure, in which proper opportunity achieves for network to sense complicated nonlinear relations in data, has been supplied. The results of proposed MLRNN were promising in comparison with the results of a Multi-Layered Feed-Forward Neural Network (MLFFNN) on the aforementioned case study.

]]> 10.1504/IJAISC.2011.042715 International Journal of Artificial Intelligence and Soft Computing, Vol. 2, No. 4 (2011) pp. 367 - 380 Kaveh Khalili-Damghani; Soheil Sadi-Nezhad Department of Industrial Engineering, Science and Research Branch, Islamic Azad University, Tehran 14778, Iran. ' Department of Industrial Engineering, Science and Research Branch, Islamic Azad University, Tehran 14778, Iran recurrent neural networks time series prediction crude oil distillation distillation capacity prediction OPEC. 2011-09-27T23:20:50-05:00 2 4 367 380 2011-09-27T23:20:50-05:00