Victor Santos - Academia.edu (original) (raw)
Papers by Victor Santos
Enzyme and Microbial Technology, 1992
A nutritional study of Xanthomonas campestris in the production of xanthan gum was carried out by... more A nutritional study of Xanthomonas campestris in the production of xanthan gum was carried out by factorial design and analysis of experiments. The concentrations of nutrients such as nitrogen, phosphorus, magnesium, and sulfur were studied by means of a factorial design at two levels, reduced to a half with four central points for error estimation. Three responses were considered, corresponding to biomass, sucrose, and xanthan concentrations, at four different times: 12, 16, 20, and 24 h. The influence of nitrogen, phosphorus, and magnesium on the biomass, and the influence of nitrogen, phosphorus, and sulfur on xanthan production, were significant, although any variable influenced sucrose consumption. An optimized medium was proposed for xanthan production.
Journal of Industrial Microbiology & Biotechnology, 2000
Experiments based on a 23 central composite full factorial design were carried out in 200-ml stai... more Experiments based on a 23 central composite full factorial design were carried out in 200-ml stainless-steel containers to study the pretreatment, with dilute sulfuric acid, of a sugarcane bagasse sample obtained from a local sugar–alcohol mill. The independent variables selected for study were temperature, varied from 112.5°C to 157.5°C, residence time, varied from 5.0 to 35.0 min, and sulfuric acid concentration, varied from 0.0% to 3.0% (w/v). Bagasse loading of 15% (w/w) was used in all experiments. Statistical analysis of the experimental results showed that all three independent variables significantly influenced the response variables, namely the bagasse solubilization, efficiency of xylose recovery in the hemicellulosic hydrolysate, efficiency of cellulose enzymatic saccharification, and percentages of cellulose, hemicellulose, and lignin in the pretreated solids. Temperature was the factor that influenced the response variables the most, followed by acid concentration and residence time, in that order. Although harsher pretreatment conditions promoted almost complete removal of the hemicellulosic fraction, the amount of xylose recovered in the hemicellulosic hydrolysate did not exceed 61.8% of the maximum theoretical value. Cellulose enzymatic saccharification was favored by more efficient removal of hemicellulose during the pretreatment. However, detoxification of the hemicellulosic hydrolysate was necessary for better bioconversion of the sugars to ethanol.
Parallel Problem Solving from Nature, 2002
SOAP (simple object access protocol) is a protocol that allows the access to remote objects indep... more SOAP (simple object access protocol) is a protocol that allows the access to remote objects independently of the computer architecture and the language. A client using SOAP can send or receive objects, or access remote object methods. Unlike other remote procedure call ...
Neural Processing Letters, 2000
This paper proposes a new version of a method (G-Prop, genetic backpropagation) that attempts to ... more This paper proposes a new version of a method (G-Prop, genetic backpropagation) that attempts to solve the problem of ¢nding appropriate initial weights and learning parameters for a single hidden layer Multilayer Perceptron (MLP) by combining an evolutionary algorithm (EA) and backpropagation (BP).The EA selects the MLP initial weights, the learning rate and changes the number of neurons in the hidden layer through the application of speci¢c genetic operators, one of which is BP training. The EAworks on the initial weights and structure of the MLP, which is then trained using QuickProp; thus G-Prop combines the advantages of the global search performed by the EA over the MLP parameter space and the local search of the BP algorithm. The application of the G-Prop algorithm to several real-world and benchmark problems shows that MLPs evolved using G-Prop are smaller and achieve a higher level of generalization than other perceptron training algorithms, such as QuickPropagation or RPROP, and other evolutive algorithms. It also shows some improvement over previous versions of the algorithm. : 68T05
This paper presents a new dynamic channel allocation technique named Simplified Maximum Packing. ... more This paper presents a new dynamic channel allocation technique named Simplified Maximum Packing. The technique provides a very good performance, near to the theoretical maximum limit appealing to the nominal channel and call reassignment concepts utilisation. Similarity with the optimal Maximum Packing strategy in conjunction with reduced complexity in terms signalling has motivated its designation. The evaluation of the optimal nominal channel pattern that match better a certain given asymmetric spatial traffic load distribution is also proposed by using two optimisation methods: the Simulated Annealing and the Genetic Algorithms. The increase of a mobile cellular system capacity and performance due to the optimal nominal channel pattern inclusion on the channel allocation technique creation is also assessed.
Genetic and Evolutionary Computation Conference, 1999
Abstract: This paper proposes a new version of a method (G-Prop-III, genetic backpropagation) tha... more Abstract: This paper proposes a new version of a method (G-Prop-III, genetic backpropagation) that attempts to solve the problem of finding appropriate initial weights and learning parameters for a single hidden layer Multilayer Perceptron (MLP) by ...
Abstrrrct-This paper presents a simulation tool aimed to assess the relation between the link qua... more Abstrrrct-This paper presents a simulation tool aimed to assess the relation between the link quality and the overall capacity on a MSS (Mobile Satellite System). This tool is particularly appropriated to compare different radio interfaces and to estimate the applicability of the UTRA radio ...
The evaluation of some prioritisation techniques based on an iterative method that allows the cal... more The evaluation of some prioritisation techniques based on an iterative method that allows the calculation of different performance parameters is presented, namely the blocking probability for new call attempts Pbl and the handover failure probability Pb2 , based uniquely on the knowledge of the terminal's mobility parameters and traffic per cell. The handover implications in the call blocking and dropping probability are assessed via simulation. The analytical description and interpretation of the results is also presented
International Work-Conference on Artificial and NaturalNeural Networks, 1999
In this paper several neural network classification algorithms have been applied to a real-world ... more In this paper several neural network classification algorithms have been applied to a real-world data case of electron microscopy data classification. Using several labeled sets as a reference, the parameters and architecture of the classifiers, LVQ (Learning Vector Quantization) trained codebooks and BP (backpropagation) trained feedforward neural-nets were optimized using a genetic algorithm. The automatic process of training and optimization is implemented using a new version of the g-lvq (genetic learning vector quantization) and G-Prop (genetic back-propagation) algorithms, and compared to a non-optimized version of the algorithms, Kohonen's LVQ and MLP trained with QuickProp. Dividing the all available samples in three sets, for training, testing and validation, the results presented here show a low average error for unknown samples. In this problem, G-Prop outperforms G-LVQ, but G-LVQ obtains codebooks with less parameters than the perceptrons obtained by G-Prop. The implication of this kind of automatic classification algorithms in the determination of three dimensional structure of biological particles is finally discused.
Fuzzy Sets and Systems, 2003
The design of fuzzy logic systems (FLS) generally involves determining the structure of the rules... more The design of fuzzy logic systems (FLS) generally involves determining the structure of the rules and the parameters of the membership functions. In this paper we present a methodology based on evolutionary computation for simultaneously designing membership functions and appropriate rule sets. This property makes it di erent from many techniques that address these goals separately with the result of suboptimal solutions because the design elements are mutually dependent. We also apply a new approach in which the evolutionary algorithm is applied directly to a FLS data structure instead of a binary or other codiÿcation. Results on function approximation show improvements over other incremental and analytical methods.
Information Sciences, 2004
Abstract This paper is focused on determining,the parameters,of radial basis function neural netw... more Abstract This paper is focused on determining,the parameters,of radial basis function neural networks (number of neurons, and their respective centers and radii) automatically. While this task is often done by hand, or based in hillclimbing methods which are highly dependent on initial values, in this work, evolutionary algorithms are used to automatically build a radial basis function neural networks,(RBF NN)
Vehicular Technology, IEEE Conference, 1998
A traffic simulator is presented along with various simulation results for an ICO and LEO type co... more A traffic simulator is presented along with various simulation results for an ICO and LEO type constellations. The simulator permits the calculation of the traffic at the satellite footprint and spotbeam level in various instances of the satellites trajectory. By using these traffic simulation results as input to a satellite resource management simulator, the fixed channel allocation (FCA) and the dynamic channel allocation (DCA) techniques were evaluated. Simulations results are presented for the various parameters in both systems
Applied Soft Computing, 2006
In this paper we present a new version of an evolutionary algorithm that finds the hidden combina... more In this paper we present a new version of an evolutionary algorithm that finds the hidden combination in the game of MasterMind by using hints on how close is a combination played to it. The evolutionary algorithm finds the hidden combination in an optimal number of guesses, is efficient in terms of memory and CPU, and examines only a minimal part of the search space. The algorithm is fast, and indeed previous versions can be played in real time on the world wide web. This new version of the algorithm is presented and compared with theoretical bounds and other algorithms. We also examine how the algorithm scales with search space size, and its performance for different values of the EA parameters.
Vehicular Technology, IEEE Conference, 2000
Several DCA techniques have been proposed in the literature over the last few years, aiming to in... more Several DCA techniques have been proposed in the literature over the last few years, aiming to increase mobile cellular system capacity. The implementation complexity of those algorithms has motivated the conceptualisation and introduction of a new channel allocation technique named simplified maximum packing which offers a similar performance to the maximum packing technique presenting however a complexity reduction in terms of involved resources and signalling. Another aspect included in this paper is the formulation of a new mathematical model to evaluate the maximum packing technique performance based on multidimensional Markov chain theory. The values resulting from the analytical model give the theoretical performance limit of a technique that uses DCA concepts and could be extrapolated as representing the simplified maximum packing technique performance. The proposed mathematical model was validated via simulation for systems with different cell layouts, interference constraints, number of available resources and traffic load values
International Conference on Universal Personal Communications, 1995
The performance of several channel assignment strategies for a LEO polar satellite constellation ... more The performance of several channel assignment strategies for a LEO polar satellite constellation is evaluated. Fixed channel allocation (FCA) and two types of dynamic channel allocation (DCA) are considered. Its advantages and disadvantages are highlighted. Moreover, the handover process implications in the call blocking and dropping probability is assessed via simulation. Strategies that are able to cope with the high number of handovers, due to the high speed of the satellites, without affecting strongly the capacity of the system are investigated. Simulation models have been developed to implement all the features evaluated including a mobility model. An analytical description and interpretation is also presented
Neurocomputing, 2000
A general problem in model selection is to obtain the right parameters that make a model fit obse... more A general problem in model selection is to obtain the right parameters that make a model fit observed data. For a multilayer perceptron (MLP) trained with back-propagation (BP), this means finding appropiate layer size and initial weights. This paper proposes a method (G-Prop, genetic backpropagation) that attempts to solve that problem by combining a genetic algorithm (GA) and BP to train MLPs with a single hidden layer. The GA selects the initial weights and changes the number of neurons in the hidden layer through the application of specific genetic operators. G-Prop combines the advantages of the global search performed by the GA over the MLP parameter space and the local search of the BP algorithm. The application of the G-Prop algorithm to several real-world and benchmark problems shows that MLPs evolved using G-Prop are smaller and achieve a higher level of generalization than other perceptron training algorithms, such as Quick-Propagation or RPROP, and other evolutive algorithms, such as G-LVQ.
International Work-Conference on Artificial and NaturalNeural Networks, 1999
... Madrid, 35 E. 23071 Jaffa (Spain) e-maih geneura@kal-el.ugr, es phone: +34 958 24 31 62 Abstr... more ... Madrid, 35 E. 23071 Jaffa (Spain) e-maih geneura@kal-el.ugr, es phone: +34 958 24 31 62 Abstract. This paper presents a new approach to the pagination prob-lem. ... The proposed approach is available in http ://kal-el. ugr. es/-j esus/layout. ...
Enzyme and Microbial Technology, 1992
A nutritional study of Xanthomonas campestris in the production of xanthan gum was carried out by... more A nutritional study of Xanthomonas campestris in the production of xanthan gum was carried out by factorial design and analysis of experiments. The concentrations of nutrients such as nitrogen, phosphorus, magnesium, and sulfur were studied by means of a factorial design at two levels, reduced to a half with four central points for error estimation. Three responses were considered, corresponding to biomass, sucrose, and xanthan concentrations, at four different times: 12, 16, 20, and 24 h. The influence of nitrogen, phosphorus, and magnesium on the biomass, and the influence of nitrogen, phosphorus, and sulfur on xanthan production, were significant, although any variable influenced sucrose consumption. An optimized medium was proposed for xanthan production.
Journal of Industrial Microbiology & Biotechnology, 2000
Experiments based on a 23 central composite full factorial design were carried out in 200-ml stai... more Experiments based on a 23 central composite full factorial design were carried out in 200-ml stainless-steel containers to study the pretreatment, with dilute sulfuric acid, of a sugarcane bagasse sample obtained from a local sugar–alcohol mill. The independent variables selected for study were temperature, varied from 112.5°C to 157.5°C, residence time, varied from 5.0 to 35.0 min, and sulfuric acid concentration, varied from 0.0% to 3.0% (w/v). Bagasse loading of 15% (w/w) was used in all experiments. Statistical analysis of the experimental results showed that all three independent variables significantly influenced the response variables, namely the bagasse solubilization, efficiency of xylose recovery in the hemicellulosic hydrolysate, efficiency of cellulose enzymatic saccharification, and percentages of cellulose, hemicellulose, and lignin in the pretreated solids. Temperature was the factor that influenced the response variables the most, followed by acid concentration and residence time, in that order. Although harsher pretreatment conditions promoted almost complete removal of the hemicellulosic fraction, the amount of xylose recovered in the hemicellulosic hydrolysate did not exceed 61.8% of the maximum theoretical value. Cellulose enzymatic saccharification was favored by more efficient removal of hemicellulose during the pretreatment. However, detoxification of the hemicellulosic hydrolysate was necessary for better bioconversion of the sugars to ethanol.
Parallel Problem Solving from Nature, 2002
SOAP (simple object access protocol) is a protocol that allows the access to remote objects indep... more SOAP (simple object access protocol) is a protocol that allows the access to remote objects independently of the computer architecture and the language. A client using SOAP can send or receive objects, or access remote object methods. Unlike other remote procedure call ...
Neural Processing Letters, 2000
This paper proposes a new version of a method (G-Prop, genetic backpropagation) that attempts to ... more This paper proposes a new version of a method (G-Prop, genetic backpropagation) that attempts to solve the problem of ¢nding appropriate initial weights and learning parameters for a single hidden layer Multilayer Perceptron (MLP) by combining an evolutionary algorithm (EA) and backpropagation (BP).The EA selects the MLP initial weights, the learning rate and changes the number of neurons in the hidden layer through the application of speci¢c genetic operators, one of which is BP training. The EAworks on the initial weights and structure of the MLP, which is then trained using QuickProp; thus G-Prop combines the advantages of the global search performed by the EA over the MLP parameter space and the local search of the BP algorithm. The application of the G-Prop algorithm to several real-world and benchmark problems shows that MLPs evolved using G-Prop are smaller and achieve a higher level of generalization than other perceptron training algorithms, such as QuickPropagation or RPROP, and other evolutive algorithms. It also shows some improvement over previous versions of the algorithm. : 68T05
This paper presents a new dynamic channel allocation technique named Simplified Maximum Packing. ... more This paper presents a new dynamic channel allocation technique named Simplified Maximum Packing. The technique provides a very good performance, near to the theoretical maximum limit appealing to the nominal channel and call reassignment concepts utilisation. Similarity with the optimal Maximum Packing strategy in conjunction with reduced complexity in terms signalling has motivated its designation. The evaluation of the optimal nominal channel pattern that match better a certain given asymmetric spatial traffic load distribution is also proposed by using two optimisation methods: the Simulated Annealing and the Genetic Algorithms. The increase of a mobile cellular system capacity and performance due to the optimal nominal channel pattern inclusion on the channel allocation technique creation is also assessed.
Genetic and Evolutionary Computation Conference, 1999
Abstract: This paper proposes a new version of a method (G-Prop-III, genetic backpropagation) tha... more Abstract: This paper proposes a new version of a method (G-Prop-III, genetic backpropagation) that attempts to solve the problem of finding appropriate initial weights and learning parameters for a single hidden layer Multilayer Perceptron (MLP) by ...
Abstrrrct-This paper presents a simulation tool aimed to assess the relation between the link qua... more Abstrrrct-This paper presents a simulation tool aimed to assess the relation between the link quality and the overall capacity on a MSS (Mobile Satellite System). This tool is particularly appropriated to compare different radio interfaces and to estimate the applicability of the UTRA radio ...
The evaluation of some prioritisation techniques based on an iterative method that allows the cal... more The evaluation of some prioritisation techniques based on an iterative method that allows the calculation of different performance parameters is presented, namely the blocking probability for new call attempts Pbl and the handover failure probability Pb2 , based uniquely on the knowledge of the terminal's mobility parameters and traffic per cell. The handover implications in the call blocking and dropping probability are assessed via simulation. The analytical description and interpretation of the results is also presented
International Work-Conference on Artificial and NaturalNeural Networks, 1999
In this paper several neural network classification algorithms have been applied to a real-world ... more In this paper several neural network classification algorithms have been applied to a real-world data case of electron microscopy data classification. Using several labeled sets as a reference, the parameters and architecture of the classifiers, LVQ (Learning Vector Quantization) trained codebooks and BP (backpropagation) trained feedforward neural-nets were optimized using a genetic algorithm. The automatic process of training and optimization is implemented using a new version of the g-lvq (genetic learning vector quantization) and G-Prop (genetic back-propagation) algorithms, and compared to a non-optimized version of the algorithms, Kohonen's LVQ and MLP trained with QuickProp. Dividing the all available samples in three sets, for training, testing and validation, the results presented here show a low average error for unknown samples. In this problem, G-Prop outperforms G-LVQ, but G-LVQ obtains codebooks with less parameters than the perceptrons obtained by G-Prop. The implication of this kind of automatic classification algorithms in the determination of three dimensional structure of biological particles is finally discused.
Fuzzy Sets and Systems, 2003
The design of fuzzy logic systems (FLS) generally involves determining the structure of the rules... more The design of fuzzy logic systems (FLS) generally involves determining the structure of the rules and the parameters of the membership functions. In this paper we present a methodology based on evolutionary computation for simultaneously designing membership functions and appropriate rule sets. This property makes it di erent from many techniques that address these goals separately with the result of suboptimal solutions because the design elements are mutually dependent. We also apply a new approach in which the evolutionary algorithm is applied directly to a FLS data structure instead of a binary or other codiÿcation. Results on function approximation show improvements over other incremental and analytical methods.
Information Sciences, 2004
Abstract This paper is focused on determining,the parameters,of radial basis function neural netw... more Abstract This paper is focused on determining,the parameters,of radial basis function neural networks (number of neurons, and their respective centers and radii) automatically. While this task is often done by hand, or based in hillclimbing methods which are highly dependent on initial values, in this work, evolutionary algorithms are used to automatically build a radial basis function neural networks,(RBF NN)
Vehicular Technology, IEEE Conference, 1998
A traffic simulator is presented along with various simulation results for an ICO and LEO type co... more A traffic simulator is presented along with various simulation results for an ICO and LEO type constellations. The simulator permits the calculation of the traffic at the satellite footprint and spotbeam level in various instances of the satellites trajectory. By using these traffic simulation results as input to a satellite resource management simulator, the fixed channel allocation (FCA) and the dynamic channel allocation (DCA) techniques were evaluated. Simulations results are presented for the various parameters in both systems
Applied Soft Computing, 2006
In this paper we present a new version of an evolutionary algorithm that finds the hidden combina... more In this paper we present a new version of an evolutionary algorithm that finds the hidden combination in the game of MasterMind by using hints on how close is a combination played to it. The evolutionary algorithm finds the hidden combination in an optimal number of guesses, is efficient in terms of memory and CPU, and examines only a minimal part of the search space. The algorithm is fast, and indeed previous versions can be played in real time on the world wide web. This new version of the algorithm is presented and compared with theoretical bounds and other algorithms. We also examine how the algorithm scales with search space size, and its performance for different values of the EA parameters.
Vehicular Technology, IEEE Conference, 2000
Several DCA techniques have been proposed in the literature over the last few years, aiming to in... more Several DCA techniques have been proposed in the literature over the last few years, aiming to increase mobile cellular system capacity. The implementation complexity of those algorithms has motivated the conceptualisation and introduction of a new channel allocation technique named simplified maximum packing which offers a similar performance to the maximum packing technique presenting however a complexity reduction in terms of involved resources and signalling. Another aspect included in this paper is the formulation of a new mathematical model to evaluate the maximum packing technique performance based on multidimensional Markov chain theory. The values resulting from the analytical model give the theoretical performance limit of a technique that uses DCA concepts and could be extrapolated as representing the simplified maximum packing technique performance. The proposed mathematical model was validated via simulation for systems with different cell layouts, interference constraints, number of available resources and traffic load values
International Conference on Universal Personal Communications, 1995
The performance of several channel assignment strategies for a LEO polar satellite constellation ... more The performance of several channel assignment strategies for a LEO polar satellite constellation is evaluated. Fixed channel allocation (FCA) and two types of dynamic channel allocation (DCA) are considered. Its advantages and disadvantages are highlighted. Moreover, the handover process implications in the call blocking and dropping probability is assessed via simulation. Strategies that are able to cope with the high number of handovers, due to the high speed of the satellites, without affecting strongly the capacity of the system are investigated. Simulation models have been developed to implement all the features evaluated including a mobility model. An analytical description and interpretation is also presented
Neurocomputing, 2000
A general problem in model selection is to obtain the right parameters that make a model fit obse... more A general problem in model selection is to obtain the right parameters that make a model fit observed data. For a multilayer perceptron (MLP) trained with back-propagation (BP), this means finding appropiate layer size and initial weights. This paper proposes a method (G-Prop, genetic backpropagation) that attempts to solve that problem by combining a genetic algorithm (GA) and BP to train MLPs with a single hidden layer. The GA selects the initial weights and changes the number of neurons in the hidden layer through the application of specific genetic operators. G-Prop combines the advantages of the global search performed by the GA over the MLP parameter space and the local search of the BP algorithm. The application of the G-Prop algorithm to several real-world and benchmark problems shows that MLPs evolved using G-Prop are smaller and achieve a higher level of generalization than other perceptron training algorithms, such as Quick-Propagation or RPROP, and other evolutive algorithms, such as G-LVQ.
International Work-Conference on Artificial and NaturalNeural Networks, 1999
... Madrid, 35 E. 23071 Jaffa (Spain) e-maih geneura@kal-el.ugr, es phone: +34 958 24 31 62 Abstr... more ... Madrid, 35 E. 23071 Jaffa (Spain) e-maih geneura@kal-el.ugr, es phone: +34 958 24 31 62 Abstract. This paper presents a new approach to the pagination prob-lem. ... The proposed approach is available in http ://kal-el. ugr. es/-j esus/layout. ...